Computer GPU/CPU/tech Mega Thread Based on new NVidia/AMD launches in 2025 and beyond. (Page 4 of 47)
|
Originally Posted By Dam17er: Respectfully, I don't think we can look at the clock speeds between two different architectures/generations and derive performance results. At this point, we don't really have any solid numbers that I'm aware of, probably because of NDA's, and I'm not relying on Nvidia's @Dam17er Thank you for your thoughtful response. To clarify, I believe that NVIDIA has effectively admitted through its own presentation that the raw performance results will be in this range. Specifically, if you look at NVIDIA's own graphs on its product page for the 5090, you will find a circa 30% increase claimed for Far Cry 6 using only ray tracing (i.e., no DLSS of any iteration). IIRC, someone who pixel counted the graphs found it to be at approximately 31%. Remember also that NVIDIA, known for its excessive marketing claims, likely cherry-picked this particular game for being higher than others. Thus, as far as raw performance, it is not looking good for at least the founder's edition card. However, it is also hard to believe that, compared to the 4090, a founder's edition 5090 card with a 28% increase in TDP and 33% less volume to put cooling in will not be thermally limited. That supports my hypothesis that board partner cards may perform better, unlike the 4090 where there was next to no difference based upon cooling (silicon lottery being the primary factor in differences between cards). Another factor that supports my hypothesis is that Jensen talked at his CES presentation extensively about how the roots of NVIDIA were in gaming and that gaming was what made AI what it is today. I believe that NVIDIA wants to throttle its competition in gaming to keep any AI competition from gaining ground. Undercutting his competition, AMD and Intel, in price and keeping board partners from investing too much in those other brands are both tactics to achieve that end. ETA: I should add that changing the fans on both front and back of the card design of the 4090 founder's edition in favor of fans only on the front for the 5090 is very helpful in serving the small form factor PC niche. I really think that this is the 5090 founder's edition's intended role. IIRC, the founder's edition review embargo for the 4090 ended one day earlier than those of 4090 board partners, an obvious effort to promote NVIDIA's own card. This might change for the 5090 if NVIDIA wants to highlight increased performance of the partner cards. |
|
Originally Posted By FLAJD11: @Lexustech48 I don't fully agree with you on the raw power of the 5090 "trucking" the 4090. One mysterious thing that has received far too limited attention IMO is the abysmal clocks of the 5090, which are actually slower than the 4090. According to NVIDIA's own website, the 5090 has a base clock of 2010 Mhz and boost clock of 2410 Mhz. Compare this with the 5080 of the same generation's improved base clock of 2300 Mhz and boost clock of 2620 Mhz. The 4090 had a base clock of 2230 Mhz and boost clock of 2520 Mhz. While the 5090 does have more CUDA cores than the 4090, 21760 versus 16384, a ~33% improvement, the clocks are really holding this back from being a better card. A net 25% to 30% improvement in raw performance is meh in my opinion. With improved clocks, it would have been closer to the ~50% improvement of last generation. I don't think that memory speed will have much effect, or at least it has not in my experience. There has thus far been no explanation as to why the 5090 base and boost clocks are down ~10% for the base and ~5% for the boost from the 5090 to 4090. It is particularly mysterious when the 5080 has clocks that are up ~8% on the base and up ~4% on the boost from the 4090. Notably on launch, the 4080 had clocks almost identical to the 4090 (with very small improvements made by the later-released 4080 Super). Has the 5090 been deliberately nerfed by clock-throttling? It seems so, but it is hard to tell. If it is instead thermally limited, why does the founder's edition use a two slot design? Is this an effort to appease board partners like Gigabyte and Asus by letting them stand out in performance with better cooling? I am sure the board partners were upset with NVIDIA being a direct competitor when the 4090 performed similarly across all variants because the 4090 founder's edition had overengineered cooling. I would note that the maximum power draw for a card using these connectors is 675w, 600 from the power cable and 75w from the PCIe rail. That means there is additional power available even with the TDP of 575w. Perhaps the plan, as hinted above, is to place NVIDIA's own card into more of a small form factor PC niche while giving board partners the more lucrative extreme performance users through improved cooling. Or maybe NVIDIA is throttling the card to offer a Titan variant in the near future. In any event, the 5090 founders edition is not as impressive as it should be in performance IMO. I also concur with your disdain for increased frame generation and other visual tricks. I don't think you are far off the mark with your observations. I think the 2 slot FE 5090 cooler will OC terribly, requiring it to be water cooled. I think the AIB cards will offer that headroom for OC in cooling. Will Nvidia mandate a nerfed BIOS for the AIB cards so they cant hit a future planned 5090TI model? It's clear that Nvidia AIB partners got through to Team Green about competition with them. Jensen probably knows that as much as they would love to make all cards in house, they do not have that capacity to meet both consumer enthusiast and business production. Heat would be my main reasoning for why the clocks on the FE 5090 are nerfed, and to be able to claim the lowest TDP of the class. The AIBs got the green light to carry more clock speed and power draw it seems. No reason to produce such beefy 4090 like coolers if they weren't planning on OC, even if the AIB stock clocks are mandated to match the FE. To revise my statement, I think the AIB 5090s will truck the AIB 4090s. The 5090 FE in stock form will likely not be as impressive generationally, but still best the 4090. |
|
Originally Posted By Lexustech48: I don't think you are far off the mark with your observations. I think the 2 slot FE 5090 cooler will OC terribly, requiring it to be water cooled. I think the AIB cards will offer that headroom for OC in cooling. Will Nvidia mandate a nerfed BIOS for the AIB cards so they cant hit a future planned 5090TI model? It's clear that Nvidia AIB partners got through to Team Green about competition with them. Jensen probably knows that as much as they would love to make all cards in house, they do not have that capacity to meet both consumer enthusiast and business production. Heat would be my main reasoning for why the clocks on the FE 5090 are nerfed, and to be able to claim the lowest TDP of the class. The AIBs got the green light to carry more clock speed and power draw it seems. No reason to produce such beefy 4090 like coolers if they weren't planning on OC, even if the AIB stock clocks are mandated to match the FE. To revise my statement, I think the AIB 5090s will truck the AIB 4090s. The 5090 FE in stock form will likely not be as impressive generationally, but still best the 4090. I think you are likely to be proven correct. If there is not meaningful increased performance from the AIB cards, I think the vast difference in card size (and likely price) from AIBs to NVIDIA's founder's edition is going to make the AIB partners look absurd, incompetent, and backward. The EVGA defection of last generation would likely be added to by more board partners. My understanding also is that the 5090 founder's edition makes changes to the orientation of internal components that likely could not be replicated by AIB partners, thus allowing the extremely slim card. I really think the 5090 founder's edition is intended primarily to fill a small form factor niche that would not otherwise be served. If it has lower performance, it will be primarily selected by that market. |
|
Originally Posted By FLAJD11: @Dam17er Thank you for your thoughtful response. To clarify, I believe that NVIDIA has effectively admitted through its own presentation that the raw performance results will be in this range. Specifically, if you look at NVIDIA's own graphs on its product page for the 5090, you will find a circa 30% increase claimed for Far Cry 6 using only ray tracing (i.e., no DLSS of any iteration). IIRC, someone who pixel counted the graphs found it to be at approximately 31%. Remember also that NVIDIA, known for its excessive marketing claims, likely cherry-picked this particular game for being higher than others. Thus, as far as raw performance, it is not looking good for at least the founder's edition card. However, it is also hard to believe that, compared to the 4090, a founder's edition 5090 card with a 28% increase in TDP and 33% less volume to put cooling in will not be thermally limited. That supports my hypothesis that board partner cards may perform better, unlike the 4090 where there was next to no difference based upon cooling (silicon lottery being the primary factor in differences between cards). Another factor that supports my hypothesis is that Jensen talked at his CES presentation extensively about how the roots of NVIDIA were in gaming and that gaming was what made AI what it is today. I believe that NVIDIA wants to throttle its competition in gaming to keep any AI competition from gaining ground. Undercutting his competition, AMD and Intel, in price and keeping board partners from investing too much in those other brands are both tactics to achieve that end. ETA: I should add that changing the fans on both front and back of the card design of the 4090 founder's edition in favor of fans only on the front for the 5090 is very helpful in serving the small form factor PC niche. I really think that this is the 5090 founder's edition's intended role. IIRC, the founder's edition review embargo for the 4090 ended one day earlier than those of 4090 board partners, an obvious effort to promote NVIDIA's own card. This might change for the 5090 if NVIDIA wants to highlight increased performance of the partner cards. @FLAJD11 right back at you, buddy  Admittedly, I just glanced at Nvidia's performance numbers and don't put too much faith in them, same as Intel's and AMD's. I pretty much just wait for independent 3rd parties to provide that data. I suppose my optimism could be coming from the recent 9800X3D launch, where before the NDA was lifted most people were thinking it was going to be an iterative improvement, but come to find out it was actually a pretty nice increase from the 7800X3D. Time will tell! I completely agree it's possible that the FE could under perform relative to board partners based on the cooling we're seeing so far. The step up from 240mm rads on 4000 series to 360mm rads on 5000 series is also telling. It would make sense to offer the FE for SFF builds and let AIBs do their thing performance wise. Again, time will tell. I wonder if the FE cards will be exclusive to Best Buy again. That was a pretty terrible move, IMO. |
|
Originally Posted By Lexustech48: I don't think you are far off the mark with your observations. I think the 2 slot FE 5090 cooler will OC terribly, requiring it to be water cooled. I think the AIB cards will offer that headroom for OC in cooling. Will Nvidia mandate a nerfed BIOS for the AIB cards so they cant hit a future planned 5090TI model? It's clear that Nvidia AIB partners got through to Team Green about competition with them. Jensen probably knows that as much as they would love to make all cards in house, they do not have that capacity to meet both consumer enthusiast and business production. Heat would be my main reasoning for why the clocks on the FE 5090 are nerfed, and to be able to claim the lowest TDP of the class. The AIBs got the green light to carry more clock speed and power draw it seems. No reason to produce such beefy 4090 like coolers if they weren't planning on OC, even if the AIB stock clocks are mandated to match the FE. To revise my statement, I think the AIB 5090s will truck the AIB 4090s. The 5090 FE in stock form will likely not be as impressive generationally, but still best the 4090. Originally Posted By Lexustech48: Originally Posted By FLAJD11: @Lexustech48 I don't fully agree with you on the raw power of the 5090 "trucking" the 4090. One mysterious thing that has received far too limited attention IMO is the abysmal clocks of the 5090, which are actually slower than the 4090. According to NVIDIA's own website, the 5090 has a base clock of 2010 Mhz and boost clock of 2410 Mhz. Compare this with the 5080 of the same generation's improved base clock of 2300 Mhz and boost clock of 2620 Mhz. The 4090 had a base clock of 2230 Mhz and boost clock of 2520 Mhz. While the 5090 does have more CUDA cores than the 4090, 21760 versus 16384, a ~33% improvement, the clocks are really holding this back from being a better card. A net 25% to 30% improvement in raw performance is meh in my opinion. With improved clocks, it would have been closer to the ~50% improvement of last generation. I don't think that memory speed will have much effect, or at least it has not in my experience. There has thus far been no explanation as to why the 5090 base and boost clocks are down ~10% for the base and ~5% for the boost from the 5090 to 4090. It is particularly mysterious when the 5080 has clocks that are up ~8% on the base and up ~4% on the boost from the 4090. Notably on launch, the 4080 had clocks almost identical to the 4090 (with very small improvements made by the later-released 4080 Super). Has the 5090 been deliberately nerfed by clock-throttling? It seems so, but it is hard to tell. If it is instead thermally limited, why does the founder's edition use a two slot design? Is this an effort to appease board partners like Gigabyte and Asus by letting them stand out in performance with better cooling? I am sure the board partners were upset with NVIDIA being a direct competitor when the 4090 performed similarly across all variants because the 4090 founder's edition had overengineered cooling. I would note that the maximum power draw for a card using these connectors is 675w, 600 from the power cable and 75w from the PCIe rail. That means there is additional power available even with the TDP of 575w. Perhaps the plan, as hinted above, is to place NVIDIA's own card into more of a small form factor PC niche while giving board partners the more lucrative extreme performance users through improved cooling. Or maybe NVIDIA is throttling the card to offer a Titan variant in the near future. In any event, the 5090 founders edition is not as impressive as it should be in performance IMO. I also concur with your disdain for increased frame generation and other visual tricks. I don't think you are far off the mark with your observations. I think the 2 slot FE 5090 cooler will OC terribly, requiring it to be water cooled. I think the AIB cards will offer that headroom for OC in cooling. Will Nvidia mandate a nerfed BIOS for the AIB cards so they cant hit a future planned 5090TI model? It's clear that Nvidia AIB partners got through to Team Green about competition with them. Jensen probably knows that as much as they would love to make all cards in house, they do not have that capacity to meet both consumer enthusiast and business production. Heat would be my main reasoning for why the clocks on the FE 5090 are nerfed, and to be able to claim the lowest TDP of the class. The AIBs got the green light to carry more clock speed and power draw it seems. No reason to produce such beefy 4090 like coolers if they weren't planning on OC, even if the AIB stock clocks are mandated to match the FE. To revise my statement, I think the AIB 5090s will truck the AIB 4090s. The 5090 FE in stock form will likely not be as impressive generationally, but still best the 4090. The blow through FE cooler design looks great to me, and at 575W, there is almost no room for AIB models to push clocks higher on a single 600W power connector. The gigantic 8-fan 7-slot AIB cards will be quieter, but are unlikely to offer significantly greater OC potential IMO. The blow through 5090 FE will likely be a disaster when mounted up against a big ass dual tower CPU heatsink though. I don't think I would run the FE unless I had liquid cooling for the CPU, or at least a pcie riser cable to relocate the GPU. |
|
Originally Posted By castlebravo84: The blow through FE cooler design looks great to me, and at 575W, there is almost no room for AIB models to push clocks higher on a single 600W power connector. The gigantic 8-fan 7-slot AIB cards will be quieter, but are unlikely to offer significantly greater OC potential IMO. The blow through 5090 FE will likely be a disaster when mounted up against a big ass dual tower CPU heatsink though. I don't think I would run the FE unless I had liquid cooling for the CPU, or at least a pcie riser cable to relocate the GPU. Small form factor cases tend to use riser cables to reverse the orientation of the GPU for this very reason of air flow, with the motherboard facing out on one side of the case and the GPU facing out on the other side (motherboard and GPU back-to-back). That is one of the reasons I hypothesize that the 5090 founders edition is intended for that small form factor use, as well as the fact that it is specifically designated as SFF-Ready by NVIDIA using its own standard. The 4090 founder's edition's fans-on-both-sides design was sub-optimal for such small form factor purposes, though it was superior for air-cooled standard cases for the reasons you describe. It was also three slots, which exceeds the SFF-Ready NVIDIA specification, even though it was the only 4090 card that could fit many small form factor cases. Also, the total wattage available is 675w with a 600w connector, as the PCIe rail can provide 75w. So there is a little room. Notably, the 4090 founder's edition appeared to be tuned as a stock card to the point where increasing power led to very little additional performance, as turning the power all the way up (133%) would allow it to use 600w but accomplish little for most cards (silicon lottery dependent). The 5090 founders edition may not be tuned to this level, meaning that a power increase (with additional cooling) in AIB cards may allow meaningful additional performance. |
|
Originally Posted By Dam17er: I wonder if the FE cards will be exclusive to Best Buy again. That was a pretty terrible move, IMO. Best Buy certainly wasn't the perfect choice, but they were massively supply constrained and at the very least Best Buy wasn't forcing you to buy bundles with an underpowered psu, slow ram, or a motherboard from the previous generation like Newegg loves to do. |
|
Originally Posted By Ayada: Best Buy certainly wasn't the perfect choice, but they were massively supply constrained and at the very least Best Buy wasn't forcing you to buy bundles with an underpowered psu, slow ram, or a motherboard from the previous generation like Newegg loves to do. Very good points. It would be nice if Microcenter was selling them as well, because their in store only model prevents bots/reduces scalpers. It's a shame how far Newegg has fallen over the years  PCWorld put up an interview with a senior director over product management teams that focuses on the FE design. Literally the 1st topic they cover is the 2 slot design and SFF builds. It really does seem like this was a focal point of their design philosophy for the FE.  Nvidia Talks RTX 5090 Founders Edition Design With the FE having 3 PCBs, and one of those being for the PCIe connector, I wonder if we'll see future GPU iterations that have a PCIe riser cable from the manufacturer specifically designed for SFF or vertical mounting. |
|
Originally Posted By Ayada: Best Buy certainly wasn't the perfect choice, but they were massively supply constrained and at the very least Best Buy wasn't forcing you to buy bundles with an underpowered psu, slow ram, or a motherboard from the previous generation like Newegg loves to do. Best Buy already has the founders listed on their website for MSRP as coming soon placeholders. Not sure if that means they are exclusive, but they are there. |
Somewhat unrelated, but Gamer's Nexus has a tear down of a prototype cooler for the FE 4090. It's a giant brick. NVIDIA's Unreleased TITAN/Ti Prototype Cooler & PCB | Thermals, Acoustics, Tear-Down |
 Absolutely Absurd RTX 50 Video Cards: Every 5090 & 5080 Announced So Far |
Become prompt critical.
|
This is interesting...24GB RTX 5080.....  
|
A tough man can take a bullet, but a wise man can dodge one. Stay focused my brothers.
|
24 fucking GB in a card that price, ugh. at least 24gb is, adequate, for most newer games on higher settings but if using ray tracing you can still hit that cap in some situations |
Twitter ID: @GattoSeh
|
Originally Posted By JTX23: Something seems up with the FE size vs the others. As much as I'd like to think they boxed up some magic it just seems off. I feel like I heard someone else say this, but I think a reasonable theory is that Nvidia could be purposely gimping the thermal performance of the FE model to appease the GPU partners. If Nvidia's model performs just as well, why spend extra on a 3rd party model outside of scarcity? Maybe EVGA leaving really did spook Nvidia a bit. |
|
MSI has circular fan technology! Truly next generation |
|
Originally Posted By MeatBag: I feel like I heard someone else say this, but I think a reasonable theory is that Nvidia could be purposely gimping the thermal performance of the FE model to appease the GPU partners. If Nvidia's model performs just as well, why spend extra on a 3rd party model outside of scarcity? Maybe EVGA leaving really did spook Nvidia a bit. I don’t know how it was with the GTX40 series, but GTX30 FE cards may as well have been vaporware. I only ever saw partner boards for sale and much more so resale; I can’t imagine FE accounted for all much production/sales. |
Member
Joined:
Jun 2022
Posts:
8198
EE: 100% (3)
IA, USA
Gamers Nexus just published a video with an Nvidia thermal engineer, so we have more info on the FE. Incredible NVIDIA RTX 5090 Founders Edition: Liquid Metal & Cooler ft. Malcolm Gutenburg |
|
Originally Posted By Dam17er: Hopefully either the MSI Suprim Liquid or the Asus Astral with the CLC. Originally Posted By Dam17er: Hopefully either the MSI Suprim Liquid or the Asus Astral with the CLC. I'm sort of leaning towards LC. I generally avoid LC unless necessary but I'm getting concerned it will be for me in this case. I jam multi-GPUs on each system to support AI workloads. The height of many of the air cooled partner cards simply won't work. Originally Posted By JustinU235: I was leaning FE 5080 but I might say "fuck it" and fo on a 5090, still FE I think. I need to watch the GN video posted above and I'm hoping it gives me more confidence on the thermals for the FE. I am not sure what to think yet. It would definitely be the goldilocks solution for my problem so I am hopeful. There is also a digits product that I think will work great for my purposes but I want to see more before I get too excited. Having a ton of vRAM is great but not the only thing that matters. |
|
Definitely going for the Suprim Liquid. That's what I have in the 4090, and really like it . Would hate the idea of dumping all of that heat inside the case. . Going to suck being at Microcenter so early..... |
And then, he arrives at the last item on his bucket list...
|
Originally Posted By JTX23: I'm sort of leaning towards LC. I generally avoid LC unless necessary but I'm getting concerned it will be for me in this case. I jam multi-GPUs on each system to support AI workloads. The height of many of the air cooled partner cards simply won't work. I've been using AIO/CLC GPUs since the 1080 ti and loath the idea of going back to air cooled. As long as your case can handle the radiator(s), I think you'll be happy. Let us know what you end up with. Originally Posted By GySgt_D: Definitely going for the Suprim Liquid. That's what I have in the 4090, and really like it . Would hate the idea of dumping all of that heat inside the case. . Going to suck being at Microcenter so early..... I also have a 4090 Suprim Liquid. The 1st one was defective, and it took a long time to get the situation rectified, but ultimately MSI gave me a very strong replacement that scores almost 1,000 points higher in Port Royal. What time do you plan on going to MC? I'm still trying to figure out timing and don't want to be there at 5:00 am. |
|
Originally Posted By Dam17er: I've been using AIO/CLC GPUs since the 1080 ti and loath the idea of going back to air cooled. As long as your case can handle the radiator(s), I think you'll be happy. Let us know what you end up with. I also have a 4090 Suprim Liquid. The 1st one was defective, and it took a long time to get the situation rectified, but ultimately MSI gave me a very strong replacement that scores almost 1,000 points higher in Port Royal. What time do you plan on going to MC? I'm still trying to figure out timing and don't want to be there at 5:00 am. I'm working 2nd shift, so getting there early will require waking up super early with little sleep. They will open at 0900, and it will be a minimal 3 hour round trip, so...... I doubt I'll get there earlier than 8. Since this thing is going to be so dang expensive, will probably just go ahead and get the MC extended warranty. Unless I'm mistaken, they will replace your defective one immediately if you go that route. Assuming that they have one to give you..... |
And then, he arrives at the last item on his bucket list...
|
Originally Posted By GySgt_D: I'm working 2nd shift, so getting there early will require waking up super early with little sleep. They will open at 0900, and it will be a minimal 3 hour round trip, so...... I doubt I'll get there earlier than 8. Since this thing is going to be so dang expensive, will probably just go ahead and get the MC extended warranty. Unless I'm mistaken, they will replace your defective one immediately if you go that route. Assuming that they have one to give you..... |
Become prompt critical.
|
Originally Posted By JustinU235: For those that do AIO GPU and AIO CPU, how do you arrange the radiators in your case? I have the cpu radiator on the top, and the gpu radiator on the side in an e-ATX case. The gpu puts out more heat by far, so possibly I will reverse that this time around. |
And then, he arrives at the last item on his bucket list...
|
Originally Posted By GySgt_D: I have the cpu radiator on the top, and the gpu radiator on the side in an e-ATX case. The gpu puts out more heat by far, so possibly I will reverse that this time around. Gotcha. I'm pretty sure my case can support front and top radiators, and I can get a side panel to support. Will have to explore the options. |
Become prompt critical.
Joined:
Nov 2023
Posts:
2647
EE: 0% (0)
USA
|
Stupid question time, a few years ago EVGA famously rage-quit the space because they said nVidia was strangling them. Yet I'm seeing "nVidia"-based cards from MSI, ASUS, Gigabyte, and half a dozen other brands that I've never even heard of before. Did nVidia single out EVGA, or back off and then refuse to deal with EVGA because of the noise, or what? |
|
Originally Posted By GySgt_D: I'm working 2nd shift, so getting there early will require waking up super early with little sleep. They will open at 0900, and it will be a minimal 3 hour round trip, so...... I doubt I'll get there earlier than 8. Since this thing is going to be so dang expensive, will probably just go ahead and get the MC extended warranty. Unless I'm mistaken, they will replace your defective one immediately if you go that route. Assuming that they have one to give you..... Good luck, man. I hope you manage to snag one. I've never purchased an extended warranty for a GPU before, but my experience with my 4090 has me considering it. Originally Posted By GySgt_D: I have the cpu radiator on the top, and the gpu radiator on the side in an e-ATX case. The gpu puts out more heat by far, so possibly I will reverse that this time around. Exactly this, although the tubing on my CPU cooler isn't long enough or flexible enough to use the side mount and keep the tubing at the bottom. Attached File I had a 420mm rad on my 13900K, and it wasn't going to fit in the case no matter what, so I just set it on top like the classy sumbitch I am. Attached File |
|
Originally Posted By Sebastian_MacMaine: Stupid question time, a few years ago EVGA famously rage-quit the space because they said nVidia was strangling them. Yet I'm seeing "nVidia"-based cards from MSI, ASUS, Gigabyte, and half a dozen other brands that I've never even heard of before. Did nVidia single out EVGA, or back off and then refuse to deal with EVGA because of the noise, or what? EVGA got sick of nvidia's bullshit and simply decided to not make GPUs anymore. We all weep because of it. |
|
Originally Posted By Sebastian_MacMaine: Stupid question time, a few years ago EVGA famously rage-quit the space because they said nVidia was strangling them. Yet I'm seeing "nVidia"-based cards from MSI, ASUS, Gigabyte, and half a dozen other brands that I've never even heard of before. Did nVidia single out EVGA, or back off and then refuse to deal with EVGA because of the noise, or what? Most of the other brands are a lot more diversified into other components. EVGA's main business was GPUs (IIRC they were the #1 or #2 producer for Nvidia), with a side business in power supplies. Sounds like they decided it wasn't profitable enough to deal with all of the effort required to produce GPUs. The video Gamer's Nexus made about the situation also mentioned that they were constantly fighting Nvidia with things they wanted to try in the design space. Nvidia have been massive dicks to partners in the past. When XFX announced they would also start producing ATI cards, Nvidia responded with "fuck you" and completely cut them off. |
Member
Joined:
Jun 2022
Posts:
8202
EE: 100% (3)
IA, USA
|
Originally Posted By Sebastian_MacMaine: Stupid question time, a few years ago EVGA famously rage-quit the space because they said nVidia was strangling them. Yet I'm seeing "nVidia"-based cards from MSI, ASUS, Gigabyte, and half a dozen other brands that I've never even heard of before. Did nVidia single out EVGA, or back off and then refuse to deal with EVGA because of the noise, or what? Most of the other companies are subsidiaries of much larger enterprises that are much more diversified. EVGA solely relied on the PC market and specifically the DIY hardware enthusiast. Over the years they tried multiple times to differentiate from the competition and were constantly told no from Nvidia. Nvidia also became much more controlling of allocation to partners with limited volume products, like Titan/Quadro cards only went to specific partners. Partners that moved millions of units got preferential treatment and more allocation of higher tier/binned products. I think the biggest factor was Nvidia making an end-around the AIBs and essentially declaring war with their FE models. |
|
Originally Posted By LordEC911: Most of the other companies are subsidiaries of much larger enterprises that are much more diversified. EVGA solely relied on the PC market and specifically the DIY hardware enthusiast. Over the years they tried multiple times to differentiate from the competition and were constantly told no from Nvidia. Nvidia also became much more controlling of allocation to partners with limited volume products, like Titan/Quadro cards only went to specific partners. Partners that moved millions of units got preferential treatment and more allocation of higher tier/binned products. I think the biggest factor was Nvidia making an end-around the AIBs and essentially declaring war with their FE models. Originally Posted By LordEC911: Originally Posted By Sebastian_MacMaine: Stupid question time, a few years ago EVGA famously rage-quit the space because they said nVidia was strangling them. Yet I'm seeing "nVidia"-based cards from MSI, ASUS, Gigabyte, and half a dozen other brands that I've never even heard of before. Did nVidia single out EVGA, or back off and then refuse to deal with EVGA because of the noise, or what? Most of the other companies are subsidiaries of much larger enterprises that are much more diversified. EVGA solely relied on the PC market and specifically the DIY hardware enthusiast. Over the years they tried multiple times to differentiate from the competition and were constantly told no from Nvidia. Nvidia also became much more controlling of allocation to partners with limited volume products, like Titan/Quadro cards only went to specific partners. Partners that moved millions of units got preferential treatment and more allocation of higher tier/binned products. I think the biggest factor was Nvidia making an end-around the AIBs and essentially declaring war with their FE models. Sucks they also stopped producing motherboards. My PC has an EVGA PSU, 3070 FTW3, and 280MM AIO. Zero problems out of any of their products. |
I'm not the one REEING, motherfucker! -FCSD2162
|
Originally Posted By Rebel31: Sucks they also stopped producing motherboards. My PC has an EVGA PSU, 3070 FTW3, and 280MM AIO. Zero problems out of any of their products. The whole situation with pisses me off to no end being a dyed in EVGA fan. |
Twitter ID: @GattoSeh
|
First video I've seen so far of someone turning off frame gen in a practical context: (spoiler alert, it's shit)  Gaming On The RTX 5090 With DLSS 4! It's incredibly disappointing to me that in 2025, $2,000 still only gets you 30 FPS at native 4k rendering resolution. No matter how much I cry about it though, that won't change reality. You will use frame gen and DLSS, and you'll learn to like it. |
|
Originally Posted By eurotrash: Just go custom water. https://www.ar15.com/media/mediaFiles/563643/IMG_8797-3435178.jpg Originally Posted By eurotrash: Originally Posted By JustinU235: For those that do AIO GPU and AIO CPU, how do you arrange the radiators in your case? Just go custom water. https://www.ar15.com/media/mediaFiles/563643/IMG_8797-3435178.jpg We could be friends 
|
|
Originally Posted By LittlePony: The whole situation with pisses me off to no end being a dyed in EVGA fan. Originally Posted By LittlePony: Originally Posted By Rebel31: Sucks they also stopped producing motherboards. My PC has an EVGA PSU, 3070 FTW3, and 280MM AIO. Zero problems out of any of their products. The whole situation with pisses me off to no end being a dyed in EVGA fan. Heh, participate in the enthusiast world long enough and every company will eventually piss you off and many eventually go away leaving you with nothing but memories and outdated hardware  .
|
Joined:
Dec 2022
Posts:
5741
EE: 0% (0)
USA
|
Originally Posted By JustinU235: RTX 5090: $1999. Twice the power of 4090.* RTX 5080: $999 RTX 5070Ti: $749 RTX 5070: $549. 4090 performance.* *Claims made by Jensen. https://www.ar15.com/media/mediaFiles/370258/Screenshot_2025-01-06_200336-3427534.jpg Availability; "For desktop users, the GeForce RTX 5090 GPU with 3,352 AI TOPS and the GeForce RTX 5080 GPU with 1,801 AI TOPS will be available on Jan. 30 at $1,999 and $999, respectively. The GeForce RTX 5070 Ti GPU with 1,406 AI TOPS and GeForce RTX 5070 GPU with 988 AI TOPS will be available starting in February at $749 and $549, respectively. The NVIDIA Founders Editions of the GeForce RTX 5090, RTX 5080 and RTX 5070 GPUs will be available directly from nvidia.com and select retailers worldwide. Stock-clocked and factory-overclocked models will be available from top add-in card providers such as ASUS, Colorful, Gainward, GALAX, GIGABYTE, INNO3D, KFA2, MSI, Palit, PNY and ZOTAC, and in desktops from system builders including Falcon Northwest, Infiniarc, MAINGEAR, Mifcom, ORIGIN PC, PC Specialist and Scan Computers." https://nvidianews.nvidia.com/news/nvidia-blackwell-geforce-rtx-50-series-opens-new-world-of-ai-computer-graphics Holy Smokes!! 
|
|
Originally Posted By Primecube: Holy Smokes!! |
|
Originally Posted By CFletch: If the 5070 actually provides 4090 performance it'd be fantastic for the price, but I doubt it Provided that it's latency is still in line with the higher tiers, it may end up being the best option for most consumers, considering that 90fps is still more than playable, and every additional frame generated above and beyond that is only really going to matter if you're using a high refresh rate display. It's a bad card, spec wise, but if everything's relying on AI generation to work well anyway, it doesn't ultimately matter that much. |