>>977640 How is x86 emulation coming along anyway? I'd like to see a phone+ barebone controller replace the need for x86 devices

>>977645 From what I'm understanding it's been coming along pretty nice and Snapdragon chips are running DaVinci: Resolve and some other software natively.

>>977640 Good I might snag the new ryzen series later towards the fall.

Windows recall killed any hype I had for the snapdragon X elite selling well enough to have a strong second generation offering by system integrators. Even if microsoft makes it opt-in as they announced, most people clocked out of that news cycle, especially as the adobe thing blew up at the same time.

>>978207 >the adobe thing I missed that one. What new jewry did they pull this time?

>>978504 >>978513

>>978504 Adobe can steal your content at any time, for any reason, and use it however they like, including to ban you from the software for which you're paying because they don't like the political content you're creating.

>>977640 I'm still using a 5600x and will probably never build another rig given how kike'd gaming has become.

These are the builds I'm comparing at the moment. AMD https://ca.pcpartpicker.com/list/CYKKrv Intel https://ca.pcpartpicker.com/list/wTTpcH Intel ITX motherboards are generally much better for a similar price. More m.2 slots, higher end chipset. Basically the only reason to go AMD is for the ability to upgrade multiple generations down the line since they have confirmed AM5 support beyond 2027. But then the question is can this cheaper AMD motherboard handle an upgrade. I don't know much about motherboards and basically all the information on the differences between them is just that different chipsets provide more IO, which is basically irrelevant on an ITX board.

https://pcpartpicker.com/list/MgTbrv Threw this together about a year ago, my old 1060 6gb STILL inside and no idea what new card to get. >4070 super for big mone, gaytracing and great efficiency >7800 xt or 7900 great famine edition for no gaytracing, more raster and less money (should probably undervolt them Don't think I'll be going 1440p any time soon, so I got half a mind to just buy a 4060 and go back to not-giving-a-shit about any of this.

>>980050 >>979957 >>979970 >>979926 The boot light, graphics card, and fans come on, but nothing else. I fell for the liquid cooling meme for my first PC and it probably crunched my CPU with a shitty alternative contact frame. Should I buy a new CPU and cooler? Just try to loosen up the contact frame? If I do, do I need to reapply the thermal paste? That little bit of thermal paste on the contact frame isn't an issue, right? Yes, the little triangles are where they're supposed to be Here's the build https://pcpartpicker.com/list/9426HG

>>980081 >>979907 >I have no idea how little time I had before it dries Thermal paste isn't glue or cement, it isn't supposed to dry. It should stay relatively "wet" for months, or years. When it dries out it becomes less effective. >If I do, do I need to reapply the thermal paste? Probably not. >That little bit of thermal paste on the contact frame isn't an issue, right? Thermal paste is non-conductive. It doesn't matter where it ends up. >but nothing else. What does "nothing else" mean? You don't get the POST screen on your monitor? Is it plugged into the GPU or the motherboard?

>>980081 >>980090 >BOOT - indicates the booting device is not detected or fail Green light in the manual says it isn't detecting a "booting device". The CPU light isn't on so it's probably detecting a CPU fine. I'll assume "booting device" means it isn't finding a bootable drive, but I'm not certain. If that was the case it should still show the POST/Bios screen.

>>980090 >You don't get the POST screen on your monitor? Is it plugged into the GPU or the motherboard? Monitor gets no signal. Thanks for the info on thermal paste. I'm going to try something else first. I think I mixed up the CPU fan and the radiator fan. The headers are labeled CPU_FAN and SYS_FAN1, but the wires are labeled FAN and VRM. I didn't realize the fan directly on the CPU was also called the VRM and had it in SYS_FAN1 with FAN in CPU_FAN. Already rewired and going to try again. >>980091 I'll check the manual again if my next attempt fails. The manuals haven't been too useful to me compared to online tutorials so far, with the exception of telling me which screws to use where.

>>980092 >I think I mixed up the CPU fan and the radiator fan. The headers are labeled CPU_FAN and SYS_FAN1, but the wires are labeled FAN and VRM. I didn't realize the fan directly on the CPU was also called the VRM and had it in SYS_FAN1 with FAN in CPU_FAN. Already rewired and going to try again. That's not going to make a difference for whether or not it will get to the POST screen. Even with shitty contact with the cooler you should still see probably at least get to the POST screen before it throttles and shuts itself off. >I'll check the manual again if my next attempt fails. I already checked the manual, green light means can't detect a booting device. Is your display cable plugged into the motherboard or the GPU?

>>980094 >Is your display cable plugged into the motherboard or the GPU? Sorry, forgot to answer this. Motherboard. I assume the GPU won't work for that until I get some drivers. >at least get to the POST screen before it throttles and shuts itself off. Doesn't seem to shut off on it's own.

>>980095 >>980094 Also it's HDMI. Gonna try DP instead.

>>980095 Swap between DP/HDMI, and swap between the GPU and the motherboard. Might need to reboot between swapping the cable around.

>>980097 Oh lmfao, I just realized you have a KF CPU. Yeah, it's because you were plugged into the MB and not the GPU. KF CPU's don't have an iGPU built into them.

>>980100 That was it. Using the the GPU after a reboot and I have visuals. Hopefully I can figure out the rest from here. Thanks anon.

>MSi CLICK BIOS 5 >Mouse works in the fucking BIOS screen Sugoi.

>>980101 In the future things like CPU coolers usually have compatible motherboard listings on their site, as well as sometimes listing compatible ram since they can sometimes hang over top of the ram. https://support.arctic.de/lf3-compatibility Their site says it is compatible with your motherboard. Though it was hidden under the "manual" button on the cooler page for some reason.

What’s the cheapest setup that can still handle all your gaming needs? If that’s not specific enough then let’s say being able to run last gen games with maximum fps without turning the settings down.

>>980108 5800X3D / 13600k + 4070S

>>980108 Were you asking for anyone specifically or just anybody. Myself I don't need much (7th gen is generally the highest end I'll go) so I'd just maximize power efficiency over everything. I lost my PC recently though so I'd have to scrape together something from scratch. https://pcpartpicker.com/list/n8DQJy Something like this would be more than enough for me.

>>980016 That cooler is already gonna struggle with a 7600, a 12700k isn't coolable with that A 12700k is strictly worse than a 13600k Don't get a non Z mobo with a K CPU or a K CPU wiith a non Z mobo

>>980118 >A 12700k is strictly worse than a 13600k More cores to dedicate for virtual machines. >Don't get a non Z mobo with a K CPU or a K CPU wiith a non Z mobo The 12700K is cheaper than the non K variant at the moment.

>>980124 >More cores to dedicate for virtual machines. 13600k has 14, the 12700k has 12 >cheaper than the non K fair enough.

>>980126 Those are E-cores, not P-cores. 12700k has 2 more P-cores.

>>980127 So? for VMs it not gonna matter unless you're doing something stupid Not like you can really choose which the VM is using either because intel is very smart

>>980129 You can pin the P cores to a VM.

>>980132 Has that gotten reliable and easy to do? Granted this might not be an issue on Linux but on windows it's painful to do last I checked

>>980133 CPU pinning is standard practice in KVM as far as I'm aware. I don't use windows so I wouldn't know. I'll probably go with AMD anyways just for a better upgrade path.

>>980127 Actually why the fuck is a 7600 an option at all if that is something you care about?

>>980135 A 7600 will work fine in a gaming VM too. I don't care particularly about min-maxing VM's, I was just making the case for a 12700k for $30 cheaper than a 13600k in terms of value. With a 7600 I lose 2 cores while gaining probably 2 more generations of CPU's to upgrade to.

>>978207 >Windows killed any hope for the snapdragon X elite selling well Fixed.

>>980016 >Basically the only reason to go AMD is for the ability to upgrade multiple generations down the line since they have confirmed AM5 support beyond 2027. The other reason is if you don't upgrade that frequently and plan to use the PC in the 2030s for AAA gaming because some games will start requiring the AVX-512 instruction set, which will be part of the PS6-era (2028) AMD CPU. For example Uncharted (2022), Dying Light 2 (2022) and Alan Wake 2 (2023) require AVX2 which was introduced in Intel Haswell / Core 4000 (2013) and AMD Zen / Ryzen (2017), even though those games could easily run on an i7-3770K if they got a programmer to do a couple days of work to add a AVX code path fallback. >>980090 >Thermal paste is non-conductive. It doesn't matter where it ends up. This isn't exactly true. Common brands of thermal paste can conduct small amounts of electricity at bursts at sufficiently high enough enough temperatures (>30C). You do not want paste/compound dripping over the CPU heatspreader because it may reach certain contacts such as the little gold circles just off the sides offIntel CPU heatspreaders to short-circuit, enough to cause random crashing. This issue doesn't happen on AMD CPUs though since they don't have those little gold circle contacts on the front.

>>980148 >This isn't exactly true. Common brands of thermal paste can conduct small amounts of electricity at bursts at sufficiently high enough enough temperatures (>30C). You do not want paste/compound dripping over the CPU heatspreader because it may reach certain contacts such as the little gold circles just off the sides offIntel CPU heatspreaders to short-circuit, enough to cause random crashing. This issue doesn't happen on AMD CPUs though since they don't have those little gold circle contacts on the front. So I should power down and find something to wipe these bits off immediately?

>>980163 If it's just on the cooler it's fine. It would have to be thermal paste with some sort of conductive mineral it, then the mineral would have to set in some way that it landed on something conductive over time. Just wipe it off with some isopropyl alcohol if it's making you nervous, with the machine turned off.

>>980137 I'm still not getting what kind of usecase with VMs would be better on a 12700k vs a 13600k but still be fine on a 7600, unless you misunderstand what kind of performance levels teach option would get you? P cores only the 13600k is slightly behind in MT vs the 12700k, but 30% faster in ST (and usually much faster in games), once you add E cores the 13600k smokes the 12700k in MT as well, both of those destroy the 7600 in both ST and MT, I'd understand comparing those to a 7800X3D (much better in games, worse with ST and MT in anything non cache sensitive)

>>980217 I mostly just remember being told the 12700k was better from VFIO nerds a year back. I could be wrong. It's also generally been cheaper than the 13600k every time I've looked, maybe slightly less so now. The 13600k was usually $100 more expensive, and even now it's $50 more expensive. A 7600 will be worse but in 8 years there will probably be some second hand 3D vcache option to do a super charged upgrade on the last generation of AM5, whereas with intel 14th gen is a massive meme. It's also $50 cheaper than the 12700k, and there was a sale that I just missed a few days ago that had it at $100 cheaper than the 12700k, and $150 cheaper than the 13600k. It's just a value analysis.

>>980065 I have a 3060 Ti and so far it's been a champ for 1080p gayman. I haven't noticed any problems from only having 8GB of VRAM and I get 90+ FPS in CP2077 @ 1080p ultra with gaytracing off. I'm planning on my next upgrade being somewhere near the end of the decade because at this point literally the only AAA game I'm planning on buying is MH Wilds and I'll be amazed if my rig can't crush it at 1080p ultra.

>>980081 >Haven't even play any games yet >Open Hydrus >Largest video file now plays instantly and I can jump to any point in the video without delay >Can zoom in on and drag around even the largest animated file while it still plays smoothly Before when I did this even on smaller animations Hydrus would lag and practically freeze. This is amazing.

>>980414 Epig

Why the fuck can't I find monitors by height? I need something 15.5 inches or smaller to fit under my upper monitor, and also 1920x1080 & 16:9 to match it. The closest I can get to regularly finding height, including the damned stand, is using the AI search on Amazon to pull info from customer reviews, and half the time that just gives me the diagonal screen size anyways. Is the customer base with limited space so damned small that something as basic as fucking product dimensions is considered a waste of time by all retailers?

>>980430 Also, monitor stands are for some reason on average much taller than TV stands. Might just get a small TV instead.

>>980430 >>980432 Nevermind. Finally found something useful. displayspecifications.com

>Small TVs that are actually available are only 720p >The very few monitors that fit under 15" tall and are at or above 24" diagonally and are actually available cost as much as, or much more than, the larger much higher quality monitor I already have

>>980489 16" tall*

>>977640 Repeating my request from last thread. I need a new network adapter but don't know what's compatible or what's a good price range. If anyone could help, I'd much appreciate it. Might also need suggestions for new speakers too. The headphone jack port on mine started acting funny about a week or two ago. I can only use it if I don't plug the cable all the way in.

>>980495 Get an M.2 A/E key to pcie 1x adapter like this: https://www.amazon.com/GLOTRENDS-Wireless-Adapter-Network-Antenna/dp/B09ZDPP43X?th=1 https://www.amazon.com/GLOTRENDS-Wireless-Adapter-Network-Antenna/dp/B09ZDPP43X?th=1 And then get any Intel AX210 m.2 card to stick on it. Used to be a lot more wi-fi vendors a few years ago, but it seems all that's left is Intel (excellent/great) Qualcomm Atheros (ok) and Realtek (hit and miss) (in terms of Linux support)

If I point a large fan at my computer, will that have any tangible effect on cooling, or will it just accelerate dust buildup?

Thanks anons. >>980584

AM4 will never fucking die. ==NEVER EVER.=÷ https://www.notebookcheck.net/AMD-Ryzen-9-5900XT-and-Ryzen-7-5800XT-AM4-desktop-CPUs-unveiled-alongside-next-gen-Zen-5-models.843490.0.html

>>980567 So would either of these work with that adapter? https://www.amazon.com/AX210NGW-5400Mbps-Tri-Band-Bluetooth-Ultra-Low/dp/B07SH6GV5S/ref=sr_1_4?crid=14N24DDMG2459 https://www.amazon.com/OKN-AX210NGW-Bluetooth-Wireless-Ultra-Low/dp/B08MJLPZPL/ref=sr_1_3?crid=14N24DDMG2459

>>977640 Anons tell me your reccomendations for a cheap gaming pc and yes I need a dvd/cd drive slot.

>>980613 How cheap?

>>980616 I dunno 700 to 900 euros max

>>980620 https://fr.pcpartpicker.com/list/YMtRWt

>>980626 Thanks

>>980630 Please actually research your shit instead of just buying a computer some random anon threw together for you. There's thousands of jewtube resources on this shit already designed for beginners.

>>980626 I'd say grab at least a 6700 XT since the 6600 is way too old, but looks like Euro prices for the 6700 XT and 7600 XT are way higher than the performance gain. In US it's 150% performance boost for 150% cost increase, and 6700 XT hits some critical breakpoints (especially with 1440p) of maintaining over 60FPS average. Euro electricity prices are supposed to be pretty absurd though. More modern GPUs will be more efficient and make up the difference in daily use.

>>980635 Thank you too

>>980648 I guess it depends on who you'll try to buy from

>>980626 Ryzen 7600 and that Asrock A620 motherboard seem like a good combo tbh. https://www.youtube.com/watch?v=NpB1c5YlO5M&t=692s

>>980648 There's too much of a budget limit for that, though >>980657 >>980630 Made a couple mistakes here https://fr.pcpartpicker.com/list/m66dRK I genuinely mixed up two mobo names, you do not want the A620M HDV/M.2 whereas the B650M HDV/M.2 is great and solid overall even with a 7950X.

>>980769 >I genuinely mixed up two mobo names, you do not want the A620M HDV/M.2 whereas the B650M HDV/M.2 is great and solid overall even with a 7950X. I'd rather have something like this instead. https://fr.pcpartpicker.com/list/LLcvJy By the time upgrading a 7600 becomes relevant, DDR6 platforms will be available.

>>980909 >DDR6 platforms will be available. I don't think Zen 6 will have it since it's AM5 for now, and probably not even whatever comes after. And to me going form a low tier Zen4 to something like a 10800X3D (if it's called that) is a significant upgrade.

>>980909 >>980769 I know this might be autistic but do these configs support win 7?

>>980939 Not really I've tried having a Zen3 build with 7 and I couldn't stop it from BSODing once a week, I can't imagine it's any better with Zen4, on Intel's side I think it's less bad but the E cores will fuck your performance into the ground on pre CPU scheduler changes Win10 (1909 I think) Honestly just consider Linux or 10 at this point (or both if you want)

Users on different forums report success with Zen 3 which shouldn't BSOD unless misconfigured. You might need to slipstream updates into your installation media. Here's an example report: >Holy crap it worked!! Thank you so much!!! <Hey, how the performance in games and such? I also want to ditch windows 10. >Well it may be different on different hardware, but it's been great. Minecraft runs at 120 fps with sildurs extreme volumetric lighting shader pack. Beam. N. G runs smooth at the highest graphics settings. I haven't tested my other games yet but it runs amazingly so far, possibly better than 10. >It may also depend on your graphics card. I have a Radeon RX580 that would bog down when spawning in multiple vehicles in Beam. N. G on windows 10, but on windows 7 I can spawn multiple cars with ease. Chrome opens a tad bit slower, which is expected, but overall I've had no problems with it.

>>980951 >Users on different forums report success with Zen 3 which shouldn't BSOD unless misconfigured I mean it seems to depend on chipset, mobo manufacturer and whether your CPU is multi CCD / CCX or not, I was trying with a 5900X on an X570 Asrock mobo Having PS/2 ports does help a lot in making it painless though. You do automatically lose 15-20% performance on multi CCD / CCX CPUs thanks to trash CPU scheduler so meh. There's also the graphics driver issue, and that's a whole lot more of an issue sicne I don't think 7 got 30 series / RDNA2 drivers

You made Zen 3 work with a 5900X on an X570 Asrock mobo so could debug occasional BSODs as you're 99% of the way to a working system. BSOD logs should be in Event Viewer to post on computer forums though integrating updates might fix it. A couple users reported better gaming performance on similar systems, suggesting downsides besides CPU scheduling may downgrade performance on 10 or 11, for example there was a VBS and HVCI issue that reduced gaming performance by 10-15% on Zen+, Zen 2, and Zen 3 on Windows 11, with functional latency increased 3X. That's an example, not the cause, the big picture probably has 10s of overhead bugs with modern Microsoft bloat. For graphics drivers some builds posted online that report running flawlessly use RTX 30 series Nvidia GPUs.

>>980909 >DDR6 platforms will be available So? DDR5 barely outperformed DDR4 for a long time. And in terms of price to performance DDR4 still shits on it as far as I'm aware. And the price to performance for being able to carry your existing DDR5 forward 4 years is massive. >>980951 >>980962 >>980973 All this autism to run windows 7 bare metal is hilarious. You could put a fraction of the effort into just learning how to pass a GPU through to a windows VM.

>>980980 >put up with Windows 10/11 in the first place Your premise is unacceptable.

>>980984 I'd imagine you can run a windows 7 VM as well, but I've never tried.

>>980980 There's often more autistic tinkering to get just one game runnning on Linux than an entire 7 install. While 10 and 11 are simplest, the conversation started when someone wanted to set up 7 so that's what it's about, it doesn't need to get derailed by more platform clannishness the same way someone asking for Christian books doesn't need to have The God Delusion shitposted at them.

If I want a pc that's optimized for ryujinx, rpcs3, and (eventually) ps4 emulation, what parts should i buy? No budget limit except "don't waste money" i don't want a part that does 100% if a part that does 90% is half price.

>>980989 >There's often more autistic tinkering to get just one game runnning on Linux than an entire 7 install. There definitely isn't in 95% of cases. So "often" is misleading. Those 5% are usually particularly annoying, though. >While 10 and 11 are simplest, the conversation started when someone wanted to set up 7 so that's what it's about The point wasn't that running games on Linux is easier than windows. Your reading comprehension is shit. The point was that running windows virtualized is easier than running it on bare metal. Perhaps try reading the posts you're responding to.

>>980992 >If I want a pc that's optimized for ryujinx, rpcs3, and (eventually) ps4 emulation, what parts should i buy? No budget limit except "don't waste money" i don't want a part that does 100% if a part that does 90% is half price. Ask communities of these emulators or contact the developers for benchmarks. Processor should be the most important. >>980998 >There definitely isn't in 95% of cases. So "often" is misleading. Those 5% are usually particularly annoying, though. It's definitely often, which isn't misleading, and definitely not 95% of cases, so "95%" is misleading. protondb's dashboard reports 5 of the top 10 Steam games don't work at all. Making one of those 50% work is possible but harder than an entire install when you'd be first to do it. From popular games that work, 43% are Tier 3 or below for tinkering, with Tier 1 being hassle-free, and 23% are Tier 4 or below. I've sat at Linux machines spending hours to make games work myself, said fuck it, and returned to 10 LTSC. Hiding this clusterfuck with fake statistics and derailing threads by calling harmless questions hilariously autistic when it's well-known Linux gaming is Autismus Maximus shit is the derailing clannishness I'm talking about.

Gaming in a VM can be more secure than gaming bare metal, though virtualization overhead introduces latency so games will be more sluggish. This can be unacceptable to many for cases like competition. Even with passthrough you won't get bare metal performance metrics and to get close takes more time than bare metal installation, though it can be faster than bare metal installation if you don't get bogged down in tweaks to minimize latency and maximize performance.

>>981001 >Hiding this clusterfuck with fake statistics Which is the exact thing you're doing right now. The top 10 games don't represent anything besides a bunch of games with invasive anti-cheat. Of the top 1000 games on steam 90% of them run. Ironically Chivalry 2 wouldn't run on windows and started bitching about EAC on Windows 7. On Linux I clicked install on the heroic launcher and it ran fine. The vast majority of games will run perfectly fine on Linux. It will be significantly more cancerous if you're trying to work with cracked games instead of GOG shit. Nothing I said is misleading. It doesn't matter if 95% of games run fine if the 5% you're trying to run require you to fiddle and fuck with shit to play them. But most anons aren't going to be playing the 5% of games that don't work to begin with. There's also been massive strides in the last 2 years. Plenty of games that always had some shit wrong with them for years can now run flawlessly. Nioh being an example of this. It's "silver" rank because it used to have broken cutscenes and such, but it actually runs essentially flawlessly with recent proton versions and has for the past 6 months. A lot of the protondb data is outdated in that regard.

>>980608 Yup, that should do it.

>>981009 No, the top 10 games represent the games people most likely play. ProtonDB Click Play, which is generally right regardless of cherrypicked outliers, rates half of those 1000 games at Tier 3 or below, which run after a shitload of tinkering because Linux gaming requires a clusterfuck of concentrated autism to get half the games to run. The autistic tinkering behind Linux gaming is easily the number one shortcoming of Linux.

>>981016 >No, the top 10 games represent the games people most likely play People are not anons. I would never tell some random normalnigger that the majority of the games they play will work fine on Linux, because they won't. I don't know why you're being so disingenuous. >rates half of those 1000 games at Tier 3 or below Already explained that a lot of that data is ancient. >The autistic tinkering behind Linux gaming is easily the number one shortcoming of Linux. Most of what people consider "autistic tinkering" is just learning how to use the operating system. Which is a massive barrier to entry, but not really a shortcoming. Yes it's extremely annoying when games don't run properly and require you to filter through some niche forum post to get a solution. But pretending like that is the case more often than not is simply an outdated experience. It'll happen more often then it will on windows, but it will not be the case with the majority of the kinds of games anons tend to play. And even for the games that do require "tinkering", a lot of it is just something you learn once and carries forward to other games. People are usually getting stumped with things that are more problems with learning the operating system, such as figuring out where wine puts the configuration files for a game.

>>981017 >People are not anons. I would never tell some random normalnigger that the majority of the games they play will work fine on Linux, because they won't. I don't know why you're being so disingenuous. It's more disingenuous to pretend random anons don't play popular games, and more disingenuous for a Linux game to derail a discussion about Windows PC hardware in the PC Hardware & News Thread to call the discussers autistic when and the clusterfuck of tinkering involved with Linux gaming is maximum autism. >Already explained that a lot of that data is ancient. No, you cherrypicked one game where it's ancient because the data isn't ancient, it's generally right. >Most of what people consider "autistic tinkering" is just learning how to use the operating system. Which is a massive barrier to entry, but not really a shortcoming. This is "It's not a bug, it's a feature" coping. Even Linus or RMS himself would take hours to get many games to work. >But pretending like that is the case more often than not More often than not? Maybe not, the numbers say half. Even at 40%, 30%, or 20% it would be extremely autistic and bad enough. Linux shouldn't be a religion and doesn't need street preachers who derail Windows discussions and conceal its shortcomings to a cultish degree. I don't do it for 10 LTSC.

>>981024 >game to derail a discussion Gamer, not game.

>>981010 Only thing holding me back is that the same seller for both is located in China but it seems like the storefronts, based on reviews, are legitimate.

>>981024 >It's more disingenuous to pretend random anons don't play popular games It's not disingenuous to assume anons don't play fucking PUBG and ASSFAGGOTS retard. >and more disingenuous for a Linux game to derail a discussion about Windows PC hardware in the PC Hardware & News Thread You're literally the one who brought up linux gaming. You responded to my post about playing games under virtualized windows with "well actually WINE is a lot of work", something my post had absolutely fucking nothing to do with. Are you legitimately fucking retarded? >No, you cherrypicked one game where it's ancient because the data isn't ancient, it's generally right. That's not what cherrypicked means. An anecdote isn't cherry picking. It's evidence of the data not accounting for recent proton updates, which applies universally to every entry dating back many years. >This is "It's not a bug, it's a feature" coping. Even Linus or RMS himself would take hours to get many games to work. I don't even know what this means. Are file systems supposed to be bugs? >Even at 40%, 30%, or 20% it would be extremely autistic and bad enough. I said even at 5% it's extremely obnoxious. I already agreed that even in the best of cases most people aren't going to stomach it over just running windows. >Linux shouldn't be a religion and doesn't need street preachers who derail Windows discussions and conceal its shortcomings to a cultish degree. Nothing about my posts were evangelically defending Linux. The disagreement was about you pretending that the majority of games today require tinkering to get running on Linux. This is false.

>>980992 >ryujinx, rpcs3 See https://www.techpowerup.com/review/amd-ryzen-7-7800x3d/16.html >and (eventually) ps4 emulation Due to lack of exclusives the incentive is too low for anyone to ever make a functional emulator for that system. PS3 (RPSC3) and Xbox 360 (xenia) are in the last generation you will see functional emulators, not including for Nintendo obviously. Instead you should be looking forward for Switch 2 emulation, which will probably need at least a 7800X3D (I'd wait for 9800X3D), 32GB DDR5 RAM, any 8GB GDDR6 VRAM GPU (e.g. RX 6600), any SSD ideally with DRAM cache.

>It's not disingenuous to assume anons don't play fucking PUBG and ASSFAGGOTS retard. Nice cherrypicking retard. >You're literally the one who brought up linux gaming. You derailed the thread by calling a couple people discussing Windows hardware autistic when you're a Linux gamer and the clusterfuck of tinkering involved with Linux gaming is maximum autism. I called out the motive for your attack and I was right because it's Linux street preachers who pull these stunts every time. >That's not what cherrypicked means. An anecdote isn't cherry picking. It's cherrypicking. You cherrypicked one game not covered right by ProtonDB out of thousands covered right by ProtonDB to say ProtonDB is wrong about the clusterfuck of tinkering involved with Linux gaming. Games in ProtonDB are generally covered right. >I don't even know what this means. It means the fault for poor compatibility is Linux's, not the user's. >I said even at 5% it's extremely obnoxious. You said 5% was annoying, not even at 5% it's extremely obnoxious, when your own top 1000 numbers show half the games are a clusterfuck of tinkering. >Nothing about my posts were evangelically defending Linux You derailed the hardware thread by calling a couple people discussing Windows hardware autistic when you're a Linux gamer and the clusterfuck of tinkering involved with Linux gaming is maximum autism, then you disingenuously concealed shortcomings of Linux for hours, because doing this is very important to you. >you pretending that the majority of games today require tinkering to get running on Linux <More often than not? Maybe not, the numbers say half. Even at 40%, 30%, or 20% it would be extremely autistic and bad enough. Linux shouldn't be a religion and doesn't need street preachers who derail Windows discussions and conceal its shortcomings to a cultish degree. I don't do it for 10 LTSC.

>>981034 >Nice cherrypicking retard. Yeah I'm not reading the rest of this. You're clearly acting in bad faith, so I'll stop enabling your derail spergout.

>>981031 >Due to lack of exclusives the incentive is too low for anyone to ever make a functional emulator for that system Aren't the current PS4 "emulators" actually just doing api translation since the PS4 is basically just a modern computer?

>>981031 Thanks but if switch 2 emulation is ever good i'll do a new pc by then, it's so far away. What about ryzen 7800 is better and how's ddr5 and dram cache help for emulation?

>>981039 SSD's without a dram cache are a miserable experience unless they're just a secondary drive. As a main OS drive you really want it to have a DRAM cache, it's worth paying the premium. SSD's without it can slug hardcore when you're downloading shit to them, sometimes to a point where they become slower than an HDD.

>>981040 I'm asking about rpcs3 and ryujinx emulation though, does ssd iops really matter for that?

>>981038 The PS4 emulator Orbital by AlexAltea uses hardware-accelerated virtualization more than emulation, but only boots the firmware, and doesn't play commercial games.

>>981039 >What about ryzen 7800 is better The closest single-thread performance equivalents to 7800X3D by Intel are i7-13700k, i9-13900k, i7-14700k and i9-14900k, but their performance is roughly the same in emulation while the Intel CPUs cost more and use more power. And if you decide to play AAA PC games the 7800X3D is a clear leader over the i9-14900k while again costing less and using less power. >and how's ddr5 Socket AM5 only supports DDR5. >dram cache help for emulation It's just my easy way of ruling out garbage SSDs as the OS drive. SSDs with DRAM are going to be high (enough) quality. Just make sure to install the latest firmware if available as soon as possible, especially with Samsung SSDs which occasionally still ship with outdated firmwares with serious premature death problems (you should still expect any SSD to die any day regardless and therefore always backup important data elsewhere like on a flash drive). Also DRAM cache is said to write-cycle longevity but it's too early to say due to lack of testing.

>>981056 >backup important data elsewhere like on a flash drive Flash drives will randomly fail more than any SSD. Backup on a secondary medium like an HDD or DVD.

I have a laptop that I'm currently turning into a NAS, it has one sata port one m.2 slot and one miniPCie which is currently populated with the wifi card. I've put a 1tb 2.5'' drive on the sata port and already filled it so I'm looking for alternatives how to upgrade the storage, I'm considering replacing the wifi card with picrel with provides two more sata sockets, but then I have no idea how to power of wire the extra drives. What do you think?

My tower came with a fan to go on the back, but the plug has three slots, and there are no headers with only 3 pins in a row on the motherboard. All the ones labeled "fan" have 4 pins. Do I need it? I haven't tried pushing the system yet, but have had not temperature issues running PCSX2 on 3x resolution and 2x speed for extended periods. There's a small fan directly on the CPU, two fans on the radiator for water cooling, and two fans on the GPU.

>>981706 invidious.private.coffee/watch?v=sVOetGYD17U

>>981723 Thank you. With this man's encouraging words, I was not intimidated.

>>981169 USB to SATA power adapters exist, you just need to stick with 2.5" drives which work off 5v alone. I'd honestly stick with USB to SATA data and power, as long as it's USB3 and get a miniPCIe to 2.5G ethernet card for faster data transfers.

I recently got a Vader 3 Pro for about $70 for its 6 extra buttons. What's are the best controllers with extra buttons for binding hotkeys to?

>>980108 A better question would be what is the cheapest setup that can run current gen vidya at 60fps. Even if you have no interest in that, if it can run current gen, it can run everything before.

>>982783 >current gen vidya >being optimized well enough to run on current hardware >not being coded by diversity hires so incompetent it would choke on Frontier TDS

>>982783 Not really complicated. Whatever the highest fps/$ config there is. So a 6700 XT with a 12400f or some shit. Probably 16GB of ram, cheapest mb, low wattage psu, basically making a computer that will be difficult to upgrade down the line. An AM4 system is also probably very cheap and capable right now.

Can someone explain how uneven ram breaks things?

>>980108 Get an Xeon CPU and then whatever cheap GPU you want. With software advances the only real limitation now is the resolution you want to run your games at. Still, it's a moot question because you are going to have shit devs that not only make games 300GB but also make them run like crap. You can find a lot of benchmark videos from turd worlders if you enter in the CPU and GPU you want to benchmark. Passmark is your friend. https://www.videocardbenchmark.net/gpu_value.html#all-time-value https://www.cpubenchmark.net/cpu_value_alltime.html

>>982858 A lot of those benchmark videos are just completely fake. I wouldn't depend on anything besides benchmarks from reputable sources.

>>981841 >and get a miniPCIe to 2.5G ethernet card for faster data transfers. Most of my client devices have gigabit ethernet, so that's not really needed. I'm also considering just getting a Direct Attached Storage and connect it to one of the USB 3.0 ports of the machine. These machines seem interesting since they are cheap compared to other alternatives: >Syba SY-ENC50104 (first pic) >Mediasonic PROBOX (second pic) They are sold for 105$ on amazon and they are non-raid, which; from my understanding they won't show as one single drive, but rather as individual units in my machine and then I can set them up with software RAID on Linux. Nevertheless, a lot of people point out the USB 3.0 connection gets faulty eventually, so I'm not sure, are USB 3.0 connections for drives reliable in the long term?. Also, there are some mixed opinions in regards of SMART, some people got it to work but some of them are getting trouble reading the data from the drives.

>>982845 Same way a car with an uneven camshaft breaks things.

>>982866 The only reliable way to hook up hard drives are ports on your motherboard (or added by your brand-name system integrator and likely has drivers) and SAS cards which can split off into sata. USB's best use case is only as a backup drive that you use once a week or month.

Reading the reply chain, since you're kind of doing a budget build its probably good enough as long as you have redundancy in software though.

>>982874 I see so do you suggest me either build a proper NAS from scratch or buy a machine from a NAS from a known vendor such as Qnap or terramaster? I also forgot to ask, a UPS would be needed to avoid data corruption on the DAS in case of a power outage? >>982875 Yeah, the laptop works just fine and it already solves all my NAS related issues aside from the storage, so I think it would be better to keep working on it rather than buying/building a dedicated device.I also think it's a fun project to see what else can be done with outdated or disused devices. >its probably good enough as long as you have redundancy in software though. I want something that could work as a place to store all my series and movies and as a torrent server. I don't really have a lot of important stuff there, but it would be a pain in the ass to rebuild my collection from scratch, so I'm planning to start with two 4TB drives and set them in RAID 1.

>>982845 Different RAM ICs (and also rank setups) have different setting they can handle / like running at, when you only have one type of ICs it's usually easy for the mobo to auto set some settings for it, when you don't it gets very confused as does your MC Most systems are now designed to run best with one or two sticks of RAM, usually people with uneven RAM have all 4 slots populated making it way worse (though two unmatched sticks can already be fairly bad.

>>982878 If you don't already have some spare desktop components to build it's probably best to look into how those vendor ones handle hdds and migrations, synology uses a proprietary filesystem which can only migrate to other systems by them, but I think can add drives with redundancy one at a time instead of a fixed number of drives at a time like zfs, for example. From there it's just figuring out how any storage upgrades would be handled, one mirrored pair to another or one pool of 4 to another pool of 4, or one device to another entirely, whatever you think you need or can afford. If you want to manage the nas os maybe a used dell r730xd with 8 3.5" drive bays can be of interest for around $500 or less, but it might be noisy, bulky, and power inefficient in comparison. Building your own out of desktop components would be about on par unless you start looking at specific parts for power efficiency. It comes down to if you want to start bothering with servers and virtual machines too or just network storage and possibly user-submitted torrent software for vendor NASs, since managing the OS yourself can be a headache. As for the UPS it's worthwhile to get one since it can maybe power your router/modem too, some rack hardware can have a battery for the write controller too but it'd consider it less reliable.

>>982761 What case did you buy?

>>982796 >Basically making a computer that will be difficult to upgrade down the line. Which would make things more expensive down the line.

>>983050 Sure, except you explicitly asked for the cheapest computer than can run existing games, not future ones. You get cheap computers by cutting corners and limiting your options. You get good computers by spending more money. If you want cheap hardware with no compromise then you hunt for sales during black friday and christmas.

>>983050 >>982796 >>983054 You can sorta make a decent compromise with current gen stuff, If you're going for a 12100 / 12400 / 7500f / 7600 / 85400f / 8500G as a base you can just put an extra 20-50 bucks on a mobo that will handle anything on the socket just fine and have a decent upgrade path With intel it's slightly less of a good compromise because there's gonna be less difference from that cheap CPU to the last flagship (14900KS) but on AMD provided you're starting from the 8400F you're going from a very meh 2+4 cores to whatever the Zen6 flagship is gonna be.

Does anyone know what the best streaming client PC would be that isn't some chink mini PC? I'm very cautious about the mini PC's flooding the market right now as they are all made by very chinky companies and some have been shipping with malware out of the box. I've heard mixed things about the new raspberry pi's. Then there's the nvidia shield devices, but those are proprietary software with microphones built into them which is annoying.

>>984486 >I'm very cautious about the mini PC's flooding the market right now as they are all made by very chinky companies and some have been shipping with malware out of the box. Just bring your own drive and install everything by yourself. I'm turning an AMD laptop into a streaming box, my OS is mint. I have a samba server and kodi configured already, only a decent YT client that could work from the TV is missing.

>>984496 >Just bring your own drive and install everything by yourself. I'm paranoid to the point where I imagine them having compromised motherboard firmware. But after doing some more research it seems like the raspberry pi 5 can handle 4K 30FPS streaming. It's just a problem of not being a licensed device for streaming from major platforms, but I don't really care about that since most of the streaming will be done from jellyfin.

>>984486 Shame you really don't want chink mini PCs because an N100 based one is pretty much what you want One of the big names has one however, but that's over 2x the price. > It's just a problem of not being a licensed device for streaming from major platforms I think the N100 is fine on that front though I might be wrong.

>>984510 >I think the N100 is fine on that front though I might be wrong. Doesn't matter if the hardware is licensed if the software isn't. You need to be running windows, and using microsoft edge browser to get anything above 720p on netflix, and presumably most other streaming services. Even if you set them to 1080p it will just output 720p without telling you. So if I need a device to stream from the nigger sites for family I'll probably just get an nvidia shield as it seems like the least cancerous of the proprietary devices.

>>984512 Just get an Amazon Fire Stick or whatever for that type of thing, and consider it separate but complimentary to your video game device.

>>984515 Why would I get a fire stick over an nvidia shield?

>>984517 Because I must have misread your intentions due to being a retard.

>>984510 >>984512 There are a couple of motherboards that come with the N100 built in, if you are willing to do your own NAS/HTPC build. Asrock has two boards with 1 M.2 SSD slot, 2 SATA connectors and a 2 lane PCIE slot (you can drop a SATA card in here for more drives)

>>984522 There's just no logical reason to go with an N100 as a 4K streaming client if a raspberry pi 5 can do 4K 30fps HDR. The raspberry pi would be like a third of the total cost in a significantly smaller form factor. I already have a sever, all I need is a cheap means of displaying the content on the TV that is 4K capable. Of course the TV's already have capable computers built right into them, but they are infested with chink spyware so they simply never get access to the network to begin with. I have a bunch of old laptops lying around but they are old enough that their GPU's can't output 4K resolution. Otherwise I'd just use those.

>>984526 The thing with the Raspberry is that they are hard to find, sometimes are overpriced and you'll have to get a case, cooling solution and power source, so once you start adding up you'll end on miniPC ranges, also you'll be using ARM which is sometimes harder to get things working compared to x86.

>>984535 They are on Amazon right now for less than $100. I don't think they are hard to find anymore. >sometimes are overpriced and you'll have to get a case, cooling solution and power source, so once you start adding up you'll end on miniPC ranges It's about $150 with a fan and PSU. Still under the miniPC range and is a known open source secure system unlike the Mini PC's. >also you'll be using ARM which is sometimes harder to get things working compared to x86. Doesn't matter for basic tasks such as being a jellyfin client.

>>984535 rpilocator.com has it in stock everywhere

>>984541 >Still under the miniPC range Nope

>>984660 The only mini PC that reaches the raspberry pi price range are used HP/lenovo machines. And the mini PC's that come close to that price are all made by the sketchiest no name chink companies possible.

>>984541 A pi5 8 gigs w/ case, cooling & AC adapter is about $130 in yuroland

>>984791 I have seen that the Pi5 does 4K60, but what are my options for 4k120/144 w/ HDR via HDMI2.1? I ask because I am running Winders 11 and I would like to go Linux full-time, but I swap displays from my monitor to my TV for various tasks, and while I really couldn't give less of a shit about gaymin in HDR (what games worth giving a shit about even support HDR?) I DO give a shit about being able to use it for proper movies. And yes, while I know that most films are 24fps, being able to output at 120 helps a LOT with stutter. I already have multiple different compact PCs kicking around, but there's not really any cheap/easy options to get a GPU to output 2.1

>>984878 I've heard HDMI 2.0 doesn't work fully in Linux due to licensing issues, and same with HDMI 2.1

>>984879 And there's not any reliable HDR implementation anywhere in Linux from my understanding, I guess KDE Plasma kind of implements it via Wayland? This shit is why I stick to Winders, as much as I hate the OS.

>>984879 Most linux-oriented devices have a hardware converter for the port for physical hdmi to a kernel-presented displayport, but obviously that's added cost and pcb real estate. I would probably wager the raspi does that if the specs match up.

anyone here know how to deal with a WHEA logger 18 Machine Check Exception error? basically I'll be playing a game and suddenly my PC will black screen and restart. I tried various things, and although the crashes have stopped for now, I don't know if I solved the problem or if I'm just stricken with temporary luck. if you've experienced this problem before, got any advice?

>>984886 If you're running a RAM OC or a CPU OC it's too strong / not juiced with enough voltage. If you're not, you have a dud CPU

>>984886 As >>984893 says it seems like the best solution is to just RMA the CPU. But I would try reinstalling the CPU, undervolting, and changing ram speeds if you haven't already. Basically everything I read about the error is wildly different. Some people are saying it's some random ryzen utility software, some are saying it's their GPU drivers. You could try running linux to troubleshoot if it's a software issue, but I doubt it actually is. But the easiest and simplest is to cry to AMD and just get a replacement.

>>984879 I think it works on Nvidia funny enough, don't quote me on that though.

>>984964 Because nvidia's shit is proprietary and isn't in the kernel. You could probably get it working with proprietary AMD driver's as well, which do exist for linux I believe.

>>984969 Ah, that makes sense then

>>984893 >>984961 thanks guys. I did some investigation myself and came to the same conclusions. might just have to go with intel this time.

>>984996 Changing platforms will be ridiculously expensive. If you have a dud CPU it should get replaced by AMD.

This month, multiple major manufacturers have been found to host sensitive customer data completely open to the internet, indexed by ordinary search engines. This includes both individual customers, and business customers like pre-built sellers. https://invidious.materialio.us/watch?v=UHxQC95syc0 https://iv.datura.network/watch?v=DeE_gLl3j94

>>988753 wtf now i love macs

Disable Accessibility Services on Firefox for less long-term memory leaks Reminder to disable Accessibility Services on both desktop PCs and Android if you don't want Firefox to hog up memory due to a serious memory leak that accumulates over time which Mozilla refuses to acknowledge. In about:config, set accessibility.force_disabled to 1 then reopen for changes to take effect. >reddit.com/r/firefox/comments/p8g5zd/why_does_disabling_accessibility_services_improve/ >reddit.com/r/browsers/comments/1d3hp8l/comment/l6a79y4/ >reddit.com/r/firefox/comments/12ijezp/comment/jfwsxq5/ >reddit.com/r/browsers/comments/10z2xk1/comment/j83dhh7/ >bugzilla.mozilla.org/show_bug.cgi?id=1726887 (closed because they "can't reproduce") >bugzilla.mozilla.org/show_bug.cgi?id=1907929

>>990133 Just did it and its crazy how everything loads instantly now.

The Intel 13th and 14th gen instability and degradation issues are becoming widespread once the people stopped doing panic fixes and started gathering data. And Intel is reported to be still fucking quiet to their fuck ups and hoping it'll blow over.

>PC Toaster Race Wins Again! Wew, Intel Rigfags BTFO'd

>>990133 Hopefully that means I don't wake up to firefox eating 16GB+ of RAM. >>990399 That crosses an intel build off the list for me, shame I kind of wanted that 12 Pcore thing they're supposedly gonna release soon.

>>990133 It’s weird that they’re called memory leaks when the problem is that memory which is no longer needed isn’t being released.

Would anyone know the potential cause of a monitor suddenly not displaying any image while still showing a backlight after being turned on? It doesn't display the monitor splash screen, doesn't display any of the on screen menu settings when the buttons are pushed, but when brightness is changed you can see it changing if you push the correct buttons in the backlight. Eventually after a random amount of on/off cycles it will show a picture, and then proceeds to show a picture perfectly fine until the next on/off cycle. It also worked fine turning on and off for about a week then starting having the problem again. Sometimes I need to cycle the power 100 times to get it to show an image. I would just replace the entire mainboard for it but there are no replacement parts or broken monitors to buy. So my only option is figuring out which surface component would cause this specific issue and soldering new ones. And I really don't feel like buying a new monitor until microLED becomes mainstream.

>>990964 What model is your monitor?

>>990964 Unplug it entirely, then plug it again. Try another cable

>>991065 AOC Q32G1WG4 >>991074 Already unplugged the power completely a few times. And the display cable doesn't matter as it doesn't even show the monitor OSD.

>>990399 > instability and degradation issues Can you summarize? I doubt I'll find any honest information otherwise.

>>991081 >And the display cable doesn't matter as it doesn't even show the monitor OSD There's a few monitors that do funky things like cut the display when an input is detected, so yes it could still matter

>>991081 >AOC Q32G1WG4 God I am so brainrotted that I saw this and thought some Qboomer was posting here. I need a vacation.

>>991088 The ratsuit latino booty is a low budget VA chink curved monitor that was being sold a few years back.

>>991083 Tl:Dr, 13th and 14th CPUs have issues, Intel went all through from blaming Nvidia, Motherboard Manufacturers, Server Providers and anyone but themselve. Every previous fix isn't guaranteed that the degradation issue is solved, nobody knows, not even Intel. https://wccftech.com/intel-14th-gen-13th-gen-cpu-instability-issues-forces-game-studio-to-switch-to-amd-advises-customers-same/ Archive taking it's time. https://archive.is/wip/Te9cm

>>991083 Also there's a long video with relevant links to the data gathered from Gamers Nexus, youtube.com/watch?v=oAE4NWoyMZk this has been going for months and Intel does not know the problem is, also reports of RMAs being denied.

>>991109 https://archive.is/Te9cm

>>991109 >brag you managed to develop a new CPU arch in record time >release with inadequate testing while tacitly encouraging mobo manufacturer to no follow the official spec >official spec is just copypasted from your previous arch so it's not even adequate and doesn't match what the chip actually does >it blows up in your face Gee who would've though.

>>990964 >>991081 >AOC Q32G1WG4 How long have you had it? I'm guessing warranty is voided now. Also guessing you tried everything from switching power and HDMI cables and AOC video guide fixes too. Can you also show take screenshots of your monitor?

>>991116 Is there any verifiable proof though beyond that? For all you know the company in question may just be doing jewish mind tricks for certain reasons.

>>991087 Switch from display port to HDMI, unplugged power from monitor, plugged back in, still no OSD or image. >>991119 >How long have you had it? I'm guessing warranty is voided now. Don't even really remember how long I've had it, a good few years now so I doubt it's in warranty since I think when I checked AOCs warrant is shorter than 2 years. >Also guessing you tried everything from switching power and HDMI cables and AOC video guide fixes too. I guess I haven't swapped out the power cable, but I doubt that would do anything. The monitor is getting power and a backlight, it's just not showing an image. And the problem is semi-intermittent. I could try none the less, I'll have to steal the power cable from my server or something.

>>991173 Yeah still no OSD or image with a different power cable.

>>991175 You're stuck gutting it and figuring out what's torched if you bent on fixing it yourself.

>>991183 I assumed that from the start, I just don't know which surface component I should start replacing as the most likely culprit. My guess is that replacing all the capacitors wouldn't be too difficult, but might not accomplish anything.

>>991109 Thank you. I'm reminded of when nVidia had a manufacturing failure with the 8800M and blamed everyone else before being forced to admit the chip had a 100% failure rate… and still wasn't forced by law to replace any purchases with gear that WASN'T broken. They were allowed to simply swap motherboards for identical chips that were also guaranteed to fail.

>>991109 I'm switching to ryzen from my 9900k soon.

>>991185 So it appears this thing is basically impossible to disassemble. Not a single visible screw, not a single video or article describing any procedures to disassemble, and all I've done is damage the plastic trying to pry the plastic shell off. Fucking AOC is a real nigger brand.

>>991175 >Don't even really remember how long I've had it, a good few years now so I doubt it's in warranty since I think when I checked AOCs warrant is shorter than 2 years. The twerps have updated their warranty to 3 years now. >Yeah still no OSD or image with a different power cable. Damn. >So it appears this thing is basically impossible to disassemble. Not a single visible screw, not a single video or article describing any procedures to disassemble, and all I've done is damage the plastic trying to pry the plastic shell off. Fucking AOC is a real nigger brand. Try these if you haven't yet, this youtube.com/watch?v=qG79dNVRWg4 or this youtube.com/watch?v=OBxOzoZQ9JU ifixit.com/Teardown/AOC+monitor+Teardown/107186 Damn screws should be either in the back or front, and then you have to pry carefully to open up the monitor.

>>991345 >The twerps have updated their warranty to 3 years now. Yeah it's been about 4 years, I checked the box. >Try these if you haven't yet, this youtube.com/watch?v=qG79dNVRWg4 or this youtube.com/watch?v=OBxOzoZQ9JU ifixit.com/Teardown/AOC+monitor+Teardown/107186 Damn screws should be either in the back or front, and then you have to pry carefully to open up the monitor. I got it open, just had to pry the fuck out of it in some specific spots. There were no screws. I'll probably just drench the motherboard in alcohol tomorrow and work up from there. I have zero clue where to buy proper surface components that won't catch fire or explode so I'll probably put off soldering new capacitors onto it.

>>991357 Better just do the usual routine of checking every damn thing in the board. Can never be too careful, and you should check for local tech repair centers, after you've check what part broke down.

>>991378 I have to imagine something as specialized as board repair where I live would cost more than it's worth in time and money to look into, especially considering the price of better monitors than this one have come down a lot. Having to spend a third of the price of a new monitor trying to fix a smeary old VA seems pointless. I'm fine with tinkering with shit myself just to see what I can accomplish or learn, but once it reaches the point of trying to get someone else to fix it it's usually just better to buy a new device entirely.

>>991172 For the quick development part there's only leaks about the bragging but you can just look at the release schedule otherwise. For the low testing it's conjecture but really you only had to actually do testing to catch those issues, any big company that did do so has a 50%+ failure to comply to stock value on new parts, this is probably the worst failure in CPU making history that I can think of and the response to it is so poor I think intel is actually damaging their brand in the long term. Tacitly encouraging mobo maker to violate the spec is easy since you just have to read the spec and note every time it says "if you do X performance increases" with barely any mention of it wrecking stability, same with it being copypasted from 12th gen as well (there's been barely any change since the early lake era despite them being fairly different in practice) As for the blowing up in their face part, well it's self evident at this point.

>>991408 Good luck man.

This CrowdStrike stuff is some Y2K shit.

>>984886 Crank up your vcore, it help stability when undervolting or overclocking

>>991736 yup

>>990964 >So my only option is figuring out which surface component would cause this specific issue and soldering new ones It's always the electrolytic capacitors. >>991357 Show some PCB pics.

Gamers Nexus on the Intel problem right now, Oxidation? youtube.com/watch?v=gTeubeCIwRw

>>991803 I'd have to take some new pictures since my reference ones are rather blurry. Drenching it with alcohol did nothing, not that I thought would accomplish much. I found a very similar logic board and PSU on ebay that look basically identical but have a slightly different serial number on them, but having to spend $50 on replacement parts feels like it would be better spent towards a new monitor. None of the capacitors looked like they were burnt or bulging, but I know little about board repair. I was going to try just blasting it with a heat gun but maybe I'll try ordering replacement capacitors and getting around to trying to actually repair it. It's a lost cause at this point anyways so it'll just sit in my garage otherwise. At the very least I could practice some soldering I suppose. I was hoping just scrubbing the contacts on the display cable might fix it.

>>991853 They don't seem to know what the problem is either since they're fishing for production codes to look for bad production dates.

What do you think about Topton Intel boards for building a NAS? I'm considering getting this one: https://www.amazon.com/KingnovyPC-Motherboard-SATA3-0-Barebone-Appliance/dp/B0BYVMNMR9

New keyboard tech https://invidious.privacydev.net/watch?v=Feny5bs2JCg

>>991803 >Show some PCB pics. There's some residual white gunk from my alcohol on some of the green PCB.

>>992341 Well blasting the shit out of the PCB's with a heat gun somehow got it to start consistently displaying the splash screen and OSD, sort of. But now there's static showing up. Haven't tested it plugged into a device yet, might blast it with a heat gun more first.

>>992373 It sounds like bad solder points to me if doing the 360 RROD fix works on it but you never know.

>>992292 What's the main difference between this and wooting?

Would it look shitty if I play games on 1080p on a 16 inches monitor with 2560x1600 resolution?

>>992438 Yeah, it would.

>>992449 Then I should look for laptops with FHD screens, thank you

>>992438 It won't look too terrible if you go for 1920x1200, but 1080p is a different aspect ratio

>>992490 I'll get to the point. I've got three laptops in my mind, these are the ones avaliable for me to buy. Acer has 2K resolution the other two are FHD, Their specs are pretty much the same but people talk shit about MSI and HP and talk good about Acer so I'm confused HP Victus 16-R1020NT MSI Katana 17 B13VGK-1030xtr Acer Predator Helios Neo 16 PHN16-71-9357 nh. Qlvey. 004

>>992400 It nullifies a key press if another press is detected. So you can hold down A and every time you press S the held A key is nullified until S is starts to travel back upwards. Enables a bunch of weird tech in games. >>992396 Yeah that what is seems like. I'll probably try another blast of heat gun on both sides of the PCB tomorrow, then if that doesn't fix it start manually trying to reflow solder joints on each of the capacitors.

>>992438 Why not game at 2560x1600 on a 16 inches monitor with 2560x1600 resolution? The Acer specs can do it.

>>992396 Not necessarily, could also be one of the power supply component getting "revived" from the heat, like the PS3 YLOD. >>992438 Yeah it'll look blurry, not unusable but blurry

>>992496 Go with either the MSI or the Acer, the HP specs look like shit if I'm reading it right online.

>>992530 >>992496 HP is trash, granted all the brand sort of are, but HP is trashiest among the trash

>>992518 That's true, guess it'll come to that. >>992528 I have 1080p monitor at home, I've tried Hitman 3 and Red Dead Redemption 2 on 720, it kind of looked like ass, I wondered if it'd be same in my case too >>992530 Their specs are all pretty much same. rtx 4070, i7 i7-14700HX, i7-13620H and 13700HX, same order. >>992531 Yeah, so I've heard, was thinking about it because it had newer cpu but people keep calling it trash

>>992517 Static went away over night but after a few power cycles the problem manifested again. So now I'll have to strip her down again and actually reflow solder joints. At this point I'm considering trying to get some leaded solder just to prevent this nigger shit from happening again.

>>991853 New update: https://www.youtube.com/watch?v=OVdmK1UGzGs tl;dw: Intel put out a press release stating that the problem is due to CPUs requesting higher voltage than intended, resulting in degradation. They state that this will be fixed via a microcode update. The next day, hidden away in a reddit comment, Intel also confirmed that there was in fact an oxidation flaw. They insist that it is completely unrelated to the instability, and that it is only responsible for some cases of instability. They also say that they were aware of the problem for at least eight months now but did not inform any customers, including those who requested RMAs and were denied. Intel has so far not stated when the oxidation problem was present in any more detail than "back in 2023". Intel also has not made any statements on potential performance decreases due to the new microcode, but have announced that it will be released in August after the Ryzen 9000 series launches. AMD has responded by delaying the release of its 9000 series into August.

>>992292 >>992400 Wooting just added it to all of their keyboards retroactively. https://yt.drgnz.club/watch?v=RxEa7k8j1Ro

cant believe you fucking low T losers are using your tech knowledge for this fucking garbage instead of producing knock off ukrainian attack drones with remove controlled pin mechanism after driving 5 hours from germany to ukraine and getting back with an old box of soviet granades you bought for the equivalent of a fucking 1050. Fuck. Its like you subhumans wanna get replaced and enslaved by fat out of shape politcians and their subhuman short brown golems

>>993917 >europoors Lol?

>>993917 Shiny

>>993921 >mystery meat mutt LOL

>>993956 Banter from euros just doesn't really hit the same. You always sound genuinely upset, sort of like women when they try to riff with men. It's just a bit embarrassing.

>>992292 >>993847 It's also been added to QMK, so pretty much any mechanical keyboard that isn't using proprietary software can enable it.

>CHYNA I mean Intel. How could these j*ws fuck up so badly? >>990964 As other anon mentioned you might try replacing all the electrolytes. A quick the Rossman's way to check which components short is to put a paper over surface mount components and launch the board for a minute, see if there are any black marks. Tbqh i have had a missing image problem long time ago, but it was related to a dying GPU which required reballing of the main chip. If you can check other cable/source, that would be a good start. >>992341 Proper 95% alcohol should not leave traces on the board. If there are leftovers, then try water/alcohol 1:1-1,5 solution and give it time to dry. Never try gasoline or any other oil thinners, they leave conducting coating on a board. Or go for pure acetone if there are no plastic parts on the board. >>992167 Don't bother with chink boards. Either only go with well known names or don't bother at all. Random and unknown errors, software incompatibilities, and so on. If you like smaller sizes, search for office nettops with 1-2 PCIe slots. One of them for extra SATA connectors, an another one for intel 2x1Gbps chip. If you aim for something more powerful, then you may try looking at aliexpress. But please consult the appropriate websites about XEON processors and boards. >>992292 >new technology absolute bullshit. shit like opposite button canceling could be easily achieved with a simple autohotkey script for a decade, if not more. Another expensive toy for numnuts. What they describe as variable press detection levels does not seem to relate to a normal mechanical keyboard at all.

>>994235 >absolute bullshit. shit like opposite button canceling could be easily achieved with a simple autohotkey script for a decade, if not more. It's new because it's being done on a hardware level, which bypasses anti-cheat which often flares up with scripts. Nullbind scripts were mentioned in the video. >Another expensive toy for numnuts. Hall effect boards are not gimmicks or toys in the slightest. They're absolutely a competitive advantage. >What they describe as variable press detection levels does not seem to relate to a normal mechanical keyboard at all. That's because it isn't a mechanical keyboard.

>>994243 >Hall effect boards >it isn't a mechanical keyboard. The cancelling they describe is not specific to a Hall effect board. It can be done with a completely normal mechanical board, and still on the actual board level, with no host-side scripts needed. My handwired, 3d-printed board could do this if I wanted it to. And not just because QMK specifically added support in response to the controversy as I mentioned in >>994172. It was possible for years if you actually programmed the firmware to do it manually (and it's just C code you can copy examples of off the internet). Since it's all handled by the board's controller and the host system only sees the key codes, there's still no anti-cheat flagging. The variable press detection does require Hall effect switches, but that part isn't necessary for the controversial cancelling.

>>994248 >The cancelling they describe is not specific to a Hall effect board Didn't say it was, but it probably has a much more intuitive implementation with them due to the nature of being able to detect a released key press instantly rather than needing to hit an actuation threshold.

>>977640 Anons tell me tour favourite weird pc cases

>>994362 Lego ones have a special place in my heart because my dad and I built one when I was a kid.

>>994362 SPHERICAL

>>994362 Some of my favorites. >>994365 These are really neat. I may have to try this some time.

>>994373 >the fucking beaver one Even better when you know it was made by a qt taxidermist. https://imgur.com/a/beaver-computer-KhxA6

AMD fucking killed it.

>>998668 AMD killed my chance at a cheap secondhand 7000 processor by releasing such trash.

>>998668 Idk if they killed it but they definitely didn't nail it.

>>998668 They killed the 9000 series being good.

>>998668 It remains to be seen if this applies to the 9900X or the 9950X since those are the flagships.

>>998668 >Better than nothing! >At least we're not Intel! A chance of Total Victory, and just like them to just botch it.

>>998668 Yeah lol, guess 12th gen intel is still the way to go.

>>998668 It seems like it might be a decent bit better for specific tasks.

>>998984 forgot pic

>>998984 >>998985 It has whatever the instructions needed are for RPCS3 now so someone should post those benchmarks. Hopefully it sells like shit and they go on sale in a few months.

>>998668 AMD never misses an opportunity to miss an opportunity.

>>998985 We need the sonic unleashed jungle joyride daytime RPCS3 test lol

>5800XT is 329 WHYYYYYYYYYYYYYYYYY

A friend of mine said you guys could help. I'm looking to get a high-end Gaming PC and my budget is $3550 CAD. I've only done pre built PCs but I wanted to dip my toes into building my own but I don't know where to start. Any help would do.

>>1004548 I just fucked around, but this might be for you. https://ca.pcpartpicker.com/list/QHyGrv

>>1004548 Got any special requirements? Like wanting to use Linux or something, I know Ryzen 6000 series chips and newer can have some problems with Linux due to it using Microsofts new security chip called Pluton which is meant to replace TPM and Intel Me/AMD PSP. Pluton also supposedly can detect pirated software so I personally avoid it. Heres something I slapped around https://ca.pcpartpicker.com/list/jnWHdH

>>1004551 >>1004566 Wow these are really crazy builds boys. My buddy who is a regular here is going to guide me through the building process. I really like the first one's case.

>>1004548 https://ca.pcpartpicker.com/list/NWbqTY https://ca.pcpartpicker.com/list/PwhWN6 An expensive one and a cheaper one

>>1004624 Why the DRAM-less Crucial P3 Plus? It's an extreme example but Digital Combat Simulator World is said to be unplayable on DRAM-less (whether NVMe or SATA) due to stuttering. Also with the size of today's AAA games (which is unwarranted on the part of today's AAA developers not bothering to compress textures and audio files but it is what it is, just another sign of industry enshittification and decay) you may as well get 4 TB if you're going to buy a top-of-the-line CPU and GPU.

>>1004773 >Why the DRAM-less Crucial P3 Plus? It's an extreme example but Digital Combat Simulator World is said to be unplayable on DRAM-les I've heard many times that DRAM less on NVMe doesn't matter quite as much, but if you say so then yeeha just upgrade to the P5 plus or an SK Hynix Platinum i don't trust anything else on the market > you may as well get 4 TB if you're going to buy a top-of-the-line CPU and GPU. There's a lot less good choices in the 4TB range, unless you're willing to go enterprise stuff which means 3x price minimum.

>>980495 https://www.amazon.com/Panda-Wireless%C2%AE-Wireless-Adapter-Antennas/dp/B0B2QD6RPX They've never failed me

>>1010302 cease reddit spacing

>>1010342 Putting the regular text directly below the link would look super awkward. That's not reddit spacing. It's proper spacing.

>>1010393 Text under a link without breaking looks fine. And spacing after a quote regardless of what you're writing always looks stupid. Neither are "proper" spacing in the context of typical imageboard formatting. Virtually every other hyperlink in this thread is posted without a space beneath or above it and they all read fine.

Are there any benchmarks out yet about zen 5 and the AVX512 support resulting in significant gains in emulation? Is it worth the premium over zen 4 processors for RPCS3?

>>1014042 >Are there any benchmarks out yet about zen 5 and the AVX512 support resulting in significant gains in emulation? Is it worth the premium over zen 4 processors for RPCS3? It's 0-10% faster, iirc a very tuned 9700X with the RDR bench is 58fps whereas a very tuned 7800X3D is 57fps RPCS3 needs AVX512 not because of the full width (512bit) instructions but of the smaller ones that aren't part of AVX2, thus a CPU being the same mostly but having massively increased full width throughput means jackshit for RPCS3, hell you can even disable 512bit instructions altogether in RPCS3 and lose almost no performance

>>1014042 >>1014159 Daily reminder that benchmarking is nothing but advertising.

>>1014213 >Daily reminder that benchmarking is nothing but advertising. Yes and no.

>>1014213 benchmarking has its place for comparison and you know, set a benchmark on general performance. Treating it as gospel however, is where noobs falter first

>>1014216 Treating them as more than a vague indication of what to expect is not a wise move >1-10% unit variance >5-10% variance from OS >5-10% variance from default settings being "optimized" >up to 10% variance from cooling variance from cooling >up to 30-40% variance from RAM between "officially supported" and "I actually know what I'm doing" settings

>>1014159 Well the price difference is much larger than 10% so I'd say that answers that question.

>>1014237 Imo if you desperately want high RPCS3 performance there's always the 12700k / 12900k, especially if you manage to grab an AVX512 capable one, or you get an engineering sample which is guaranteed to have AVX512 at the cost of nerfing your GPU PCIe bandwidth which is also a big deal in some games on RPCS3 (but usually only on lighter games at very high framerates) or you can go for the full pro gamer move and grab a BGA to LGA 12900HX ES, which has AVX512, full PCIe 4.0 bandwidth and is under $200, and with a bit of know how will perform within 10% of the desktop chip

>>1014247 >Imo if you desperately want high RPCS3 performance I don't really care that much, I was just curious if there was a substantial uplift since no CPU reviewers ever cover emulation performance. I'll probably go AMD for multi-generation support, especially since everything after 12th gen intel is now confirmed to be completely defective.

>>1014254 i bought a 7950x3d (overpaid by probably $80 on amazoon) the processor is great, i'm loving it so far but i would definitely recommend for multigeneration support; LGA1700 is EOL soon and there's no upgrade path since intel just keeps changing their sockets. AMD doesnt have this issue, as they still fully support AM4 motherboards almost 8 years past their inception and introduction into mainstream computers :P intel is a good budget option but they dont have AVX512 support unless you buy an old production sample (basically impossible since we're now 2 generations in) and also doesnt run much faster than a zen3 (ryzen 5000 series) chip while also having tiny L3 cache in comparison. its just an overall downgrade from AMD's chips now. intel USED to be a sensible and good option, but they've fallen from grace. AM5 sockets are a bit pricier but thats always how its been for AMD motherboards. i paid a fat 400 dollars for a taichi carrara and i dont regret it a single bit. you can easily run a B650 motherboard with a top of the line AMD CPU and not have any issues though, so its by far the most flexible when it comes to cost. go AMD! (FYI the 7800x3d is selling for around $319 and the 7950x3d is selling for about $429 USD. dont pick up a 7600x3d or 7900x3d because they're worse in performance than the two previously mentioned ones. happy hunting!!)

>>1014268 I see it's your first time on an imageboard.

shut the fuck up

>>1014254 Right now if you want the best allrounder without too much messing around the 7800X3D probably is the choice >especially since everything after 12th gen intel is now confirmed to be completely defective. Well 13th and 14th , 15th hasn't been rushed to hell and is on TSMC N3 which is quite widespread already so it might actually not be a fucking nightmare

>>1014268 Been eyeing the 7950x3d for a while now might snag it during nog Friday.

>>1014278 Anything beyond a 7600X just seems like terrible value. My plan was to go 7600X this fall/winter then buy whatever the best value secondhand CPU is on that platform in like 10 years.

>>1014280 AM5 in general seems like a terrible value rn, DDR5 ram hasn't reached it's peak memory timings like Zen 2/3 has, give it perhaps one more gen

>>1014281 Yes, but the convenience of having potentially 8 years of upgrades on the used market on a single platform feels like it claws back that value.

>>1014280 The 7600X is poor value when the 7500F exists

>>1014293 I need an iGPU.

>>1014279 You should check the latest benchmarks for both 7800X3D and 7950X3D, watched and read people preferring the 7800X3D.

>>977640 if my old chipset can run doom, why would i want this one?

>>1014574 Do you play nothing except Doom?

>>1015259 Yes.

>>1015276 >nu-doom and cuckchan template meme

>>1015276 Why are we getting so many shitty normalnigger memes being posted lately?

>>1015583 Gigachad started out as an imageboard meme

>>1015587 Was that imageboard cuckchan?

How the fuck do you even deal with HDD noises. I got like 5 drives now on a docking bay and they're driving me nuts. >inb4 just get SDD bro But I need to haord

>>1018284 you know you can soundproof them but you'll have to ventilate too

>>1018284 Server/NAS, stick in a closet or other room in the house.

>>1018284 The same way you dealt with it for the last 30 years. You're not… underage, are you?

>>1018284 How fucking old are your HDDs? Mine barely make a whisper.

>>1018355 I don't hear my HDDs either.

>>1018284 Low power optimizations, maintenance diagnostics during daytime, lots of ram for caching. Anything beyond that is a physical case problem or crappy drives, using rubber dampers might cause even more damage since it starts counter-shaking. If you're using some quick reliece plastic instead of screws it could cause noise. Beyond that anything underneath it is essentially a sound amplifier, putting it on a dresser, shelf, or hardwood floor will really rumble so I have a little tower pc cart. Class 0 fireproof sound insulating foam would help soundproof your rack if you have one but I don't remember what type is good.

>>1018284 offload your OS and primary programs to an SSD, you can get pcie 3.0 2tb drives for around $100 if you shop around. Then set your HDD's to turn off when not in use. Both windows and linux have a timeout features for HDD's to turn off after X minutes. If you are just data haording 99% of the time you aren't accessing most of your collection so they will be off.

>>1019502 Hard drives are meant to spin then stay in spin, if you keep stopping then starting the spin it can damage the drive.

>>1019844 0/10, no one will believe you.

>>1019881 Bro, that's how green drives work, they're made to fail.

>>1020050 So are your lungs. Your heart. Cock and balls. The human brain.

>>1020076 But what about bones?

>>1019844 DVDs break when they aren't spinning too.

>>1020085 Fuck I have DVDs that haven't spun in decades. Is there even any point checking their status or should I just throw the lot in the bin?

>>1020089 I've had plenty of Japanese brand DVD-Rs die so I wouldn't know, people like to say that pressed discs are better so they can keep their shrink wrapped investments.

>>1020093 Ok, but what if they make special DVD cases with a small engine that spins the DVD even when it's stored on a shelf. To use with your most valuable DVDs, like the Special Extended Editions of The Lord of the Rings Trilogy.

>>1020085 How does that work?

>>1020089 Why have DVDs when you can have piracy?

Supposedly the 5080 and 5090 specs have leaked. https://archive.ph/CkKJx 90: 600W TDP, 32GB vRAM 80: 400W TDP; 16GB vRAM I say again: what a fucking shitshow.

>>1022002 On-chip optical interconnects when?

>>1022008 >optical interconnects Not gonna happen until their is a massive paradigm shift with how processors are made. We will get graphene before we get optical

>>1022028 *there is

>>1020123 the media they are made of deteriorates over time, especially if they are exposed to sunlight

>>1022002 it only seems like it from a non enterprise user perspective. Never forget that what we get is the trash chips that didn't make the cut for enterprise level systems. nvidia makes all their stuff around big computing, servers and AI now. They are only keeping a foot in the door for gaming so their stock price doesn't go down but they have more or less relegated gaming to an afterthought.

>>1022002 What's the shitshow, just the wattage?

>>1022061 16GB vram is what 2020 consoles had. To put it on a halo priced card in 2025 is hilarious.

Anyone else catch wind of this? Is it true? Bullshit? Sounds a little too convenient that everything would hinge on one mine, especially when chips use only a small portion of silicon anyways. Archive: https://archive.ph/fltGO

>>1022503 I take it you didn't hear what Russia just pulled: >>>/pol/21871

>>1022503 nah, thats bullshit, its the same claim as taiwan being the ONLY place in the world for microchips or some bullshit like that

>>1022506 >its the same claim as taiwan being the ONLY place in the world for microchips Anon, Taiwan IS the only place to develop microchips. Not because it's impossible to build them anywhere else (Like we did in the past) but because they're the only country with the infrastructure to do it. Attempting to build (Or even restart) that industry is going to require a huge amount of time and money sunk to make the infrastructure viable, so most companies don’t bother and just have the Taiwanese take care of the work. Yes, this makes things much more unstable and risky. That’s the point.

>>1022504 What's the takeaway from this in your perspective? That Russia has placed these limits in retaliation against Western sanctions, and the West is manufacturing a shortage to make war with Russia seem more viable?

>>1022511 >That Russia has placed these limits in retaliation against Western sanctions, and the West is manufacturing a shortage to make war with Russia seem more viable? That's one way to look at it that I see as agreeable.

>>1022512 How else could it be construed?

>>1022517 That they are trying to artificially inflate the value of basic resources that they sell while building up a stockpile of said resources to dump on the market when the price goes up. The only way that doesn't work is if some overseas competitor ramps up production of the same resources to steal market share from the Russian mining companies. The gambit is a bit of a gamble, but it makes sense that Russia would need to somehow brighten future income prospects in the face of the skyrocketing debt that is resulting from the enormous cost needed to modernize their international joke of a military.

>>1022002 >>1022065 >16 gb of vram The fucking 7600 XT, has 16 GB. What a joke

>>1022538 >The gambit is a bit of a gamble, but it makes sense that Russia would need to somehow brighten future income prospects in the face of the skyrocketing debt that is resulting from the enormous cost needed to modernize their international joke of a military. Only need a few more billions for Ukraine for that. Like wew, the imageboards have been infected with Yuropoor braintrot like /k/ has, with a huge hard on for anything NATO made, even their shit policies.

>>1022602 >the imageboards have been infected with Yuropoor braintrot like /k/ has But I do live in Eastern Europe in a country bordering Ukraine, so from my point of view, America is the lesser evil, and for my benefit, they should do everything in their power, short of nuclear Armageddon, to weaken Russia by proxy of Ukraine.

>>1022602 More like you have swallowed gopnik propaganda from a hose, anon.

>>1022602 >Russia stronk! See how they fight ALL OF NATO off all by themselves! <as they sit in the same trenches for two years salvoing Iranian drones at just Ukraine who is using Soviet and 20th century American surplus equipment

>>1022603 I live in america and think you need to contribute to the nato budget more, and pay us more for tanks and planes to defend yourself better. >>1022538 I hope this means shitty suicide drones cost 10x to make.

>>1022641 >I live in america and think you need to contribute to the nato budget more, and pay us more for tanks and planes to defend yourself better. Fair enough.

>>1022603 How about you pay for your own defense, coward. NATO is unconstitutional and we're bankrupt. You'll be part of Russia when the US collapses. >>1022609 Back to reddit, yid. >>1022633 >NATO is so strong! See how they lose to Russia on its own for two years straight! lol

>>1022647 >beingthis retarded I assume that anon is a pole, poland spends 4% of its GDP on defense, and all of Europe in eneral has been ramping up its expending for the past years, but because youre retarded and just read headlines, you will never actually inform yourself before talking shit.

No one gives a shit about Slavic shit holes.

>>1022503 >Sounds a little too convenient that everything would hinge on one mine No, having only one source for a key material isn't convenient at all, because it means it's vulnerable to natural disasters like this. It reminds me of how, for many decades, aluminum production hinged on a mineral mined from a single deposit in Greenland, but the mine ran dry and now the mineral is no longer mined anywhere. Aluminum production has to first synthesize it instead.

>>1022653 Nothing you say contradicts what he's saying.

Will the 5090 be a ripoff?

>>1023455 I would be surprised if it wasn't

>>1023455 VERY small chiwanese peasantcompany. Please understand.

>>1023455 Everything since the 10 series was a ripoff.

>>1022065 >16GB vram is what 2020 consoles had. PS5 has 16 GB of unified RAM and 3.5 GB is reserved for the OS, so that leaves 12.5 GB for games to use as RAM or VRAM. Since the typical AAA game today uses around 4~6 GB of RAM for non-graphics stuff (mostly generalized game engine bloat frankly but that's another topic), you're really only left with about 8 GB to use as "VRAM" (to store textures mainly, and shaders to a lesser extent). PS5 GPU is equivalent to 8 GB cards like the RX 5700 XT, RX 6650 XT and RTX 3060. On a side note, PS5 Pro GPU is speculated to be equivalent to RX 6700 XT and RTX 4060 Ti (8GB).

>>1023602 >PS5 GPU is equivalent to 8 GB cards like the RX 5700 XT, RX 6650 XT and RTX 3060. On a side note, PS5 Pro GPU is speculated to be equivalent to RX 6700 XT and RTX 4060 Ti (8GB). Physically the PS5 is closest to a 6700 (it's RDNA1 with RDNA2 RT spliced in, hard to say which exact impact that has), the PS5 Pro is closest to a 7800XT (though again it has RDNA4 RT spliced in, so it's probably way better in RT), though obviously they cannot use the same TDP (roughly guesstimating it's probably around 120-140W dedicated to the GPU part)