8GB VRAM not enough (and 10 / 12)

Page 10 - Seeking answers? Join the AnandTech community: where nearly half-a-million members share solutions and discuss the latest tech.

BFG10K

Lifer
Aug 14, 2000
22,709
2,979
126
8GB
Horizon Forbidden West 3060 is faster than the 2080 Super despite the former usually competing with the 2070. Also 3060 has a better 1% low than 4060 and 4060Ti 8GB.
Resident Evil Village 3060TI/3070 tanks at 4K and is slower than the 3060/6700XT when ray tracing:
Company Of Heroes 3060 has a higher minimum than the 3070TI:

10GB / 12GB

Reasons why still shipping 8GB since 2014 isn't NV's fault.
  1. It's the player's fault.
  2. It's the reviewer's fault.
  3. It's the developer's fault.
  4. It's AMD's fault.
  5. It's the game's fault.
  6. It's the driver's fault.
  7. It's a system configuration issue.
  8. Wrong settings were tested.
  9. Wrong area was tested.
  10. Wrong games were tested.
  11. 4K is irrelevant.
  12. Texture quality is irrelevant as long as it matches a console's.
  13. Detail levels are irrelevant as long as they match a console's.
  14. There's no reason a game should use more than 8GB, because a random forum user said so.
  15. It's completely acceptable for the more expensive 3070/3070TI/3080 to turn down settings while the cheaper 3060/6700XT has no issue.
  16. It's an anomaly.
  17. It's a console port.
  18. It's a conspiracy against NV.
  19. 8GB cards aren't meant for 4K / 1440p / 1080p / 720p gaming.
  20. It's completely acceptable to disable ray tracing on NV while AMD has no issue.
  21. Polls, hardware market share, and game title count are evidence 8GB is enough, but are totally ignored when they don't suit the ray tracing agenda.
According to some people here, 8GB is neeeevaaaaah NV's fault and objective evidence "doesn't count" because of reasons(tm). If you have others please let me know and I'll add them to the list. Cheers!
 
Last edited:

Stuka87

Diamond Member
Dec 10, 2010
6,240
2,559
136
But how can we differentiate between games that dynamically use (not necessarily need) vram according to the cards capacity of vram and games that actually will not run properly due to the lack of vram? And not confuse the latter with an already weaker GPU due to its other specs?

It requires testing on a card with less VRAM that is still fast enough to run the game at the required settings.

So if somebody ran the test with a 10GB 3080, and then a 20GB 3080 (rare, but they exist), they would see the 10GB model have much worse performance when it ran out of a VRAM compared to the 20GB model.

They would also see system RAM usage start to climb on the 10GB model as the system memory was having to be used to data that is being shuffled in and out of the VRAM.
 

Golgatha

Lifer
Jul 18, 2003
12,682
1,598
126
It requires testing on a card with less VRAM that is still fast enough to run the game at the required settings.

So if somebody ran the test with a 10GB 3080, and then a 20GB 3080 (rare, but they exist), they would see the 10GB model have much worse performance when it ran out of a VRAM compared to the 20GB model.

They would also see system RAM usage start to climb on the 10GB model as the system memory was having to be used to data that is being shuffled in and out of the VRAM.

At the 12GB level, a comparison between the 3080 Ti and 3090 would be pretty telling, as they're only about 3% difference in performance under less than 12GB VRAM usage.
 

amenx

Diamond Member
Dec 17, 2004
4,007
2,276
136
It requires testing on a card with less VRAM that is still fast enough to run the game at the required settings.

So if somebody ran the test with a 10GB 3080, and then a 20GB 3080 (rare, but they exist), they would see the 10GB model have much worse performance when it ran out of a VRAM compared to the 20GB model.

They would also see system RAM usage start to climb on the 10GB model as the system memory was having to be used to data that is being shuffled in and out of the VRAM.
I know there are games, scenarios where less vram will hobble performance. And the best way to demonstrate that are identical cards, specs, but one with double the vram. Have seen such tests but they are not many (RX 6500 4 vs 8gb, 3060 12 vs 3070) and a few others. But vast majority of games to date where ppl rant most about vram (mainly due to monitoring vram allocated) performance is quite often not affected. Thats the point I wish to make.
 
Reactions: TESKATLIPOKA

amenx

Diamond Member
Dec 17, 2004
4,007
2,276
136
Another game where a lot was made about insufficient vram hobbling performance was FC6 with HD texture pack at 4k (and where AMD stated 16gb minimum was required to run it).

 
Reactions: Heartbreaker

BFG10K

Lifer
Aug 14, 2000
22,709
2,979
126
Another game where a lot was made about insufficient vram hobbling performance was FC6 with HD texture pack at 4k (and where AMD stated 16gb minimum was required to run it).
LOL, what are you babbling about? Far Cry 6 is shown right there in the OP.


1% low FPS: 9 vs 56. Wait don't tell me...it's the game's fault for daring to exceed 640KB 8GB, and it's the reviewer's fault because he doesn't have permission to test that area at those settings, amrite?

I just love the way you innocently popped into the thread telling us "oh, I'm not really sure how to tell I've run out of VRAM, please help explain it to me!"

Here's a very simple hint: when a 3060 trashes a 3080, it's not the game's fault, not the reviewer's fault, not AMD's fault, not the player's fault, and not any other nonsensical apologist handwaving we're given each time another game shows issues.

OP now shows ten different games. Every single one is nVidia's fault, because nVidia continues to make new GPUs with specs from the year 2014.
 
Last edited:

amenx

Diamond Member
Dec 17, 2004
4,007
2,276
136
LOL, what are you babbling about? Far Cry 6 is shown right there in the OP.


1% low: 9 FPS vs 56 FPS. Wait don't tell me...it's the game's fault for daring to exceed 640KB 8GB, and it's the reviewer's fault because he shouldn't be testing that area, amrite?

I just love the way you innocently popped into the thread telling us "oh, I'm not really sure how to tell I've run out of VRAM, please help explain it to me!"

Here's a very simple hint: when a 3060 trashes a 3080, it's not the game's fault, not the reviewer's fault, not AMD's fault, not the player's fault, and not any other nonsensical apologist handwaving we're given each time another game shows issues.

OP now shows ten different games. Every single one is nVidia's fault, because nVidia continues to make new GPUs based on specs from the the year 2014.
I dont watch video benchmarks, much prefer websites with static charts that explain it. And did it skip your awareness that I was referencing the same site, video channel? HUB = Techspot!

None of that was mentioned in the TS review, in fact it leads one to an entirely different conclusion!

 

BFG10K

Lifer
Aug 14, 2000
22,709
2,979
126
I dont watch video benchmarks, much prefer websites with static charts that explain it.
Static charts like this one, from the same video?


None of that was mentioned in the TS review, in fact it leads one to an entirely different conclusion!
So what you're saying is the video from HWUB is a fabricated lie? Or are you saying the video doesn't actually exist and is a figment of our collective imagination?
 
Last edited:

amenx

Diamond Member
Dec 17, 2004
4,007
2,276
136
So what you're saying is the video from HWUB is a fabricated lie? Or are you saying the video doesn't actually exist and is a figment of our collective imagination?
I'm saying that HUB are offering different conclusions from what they stated earlier in their TS site. I would like to see the same replicated conclusion from other sites that dont contradict themselves.
 

BFG10K

Lifer
Aug 14, 2000
22,709
2,979
126
I'm saying that HUB are offering different conclusions from what they stated earlier in their TS site. I would like to see the same replicated conclusion from other sites that dont contradict themselves.
This is simply comical beyond belief. You...you do understand that testing the game in one area might show different results when tested in another area?
 

Stuka87

Diamond Member
Dec 10, 2010
6,240
2,559
136
I'm saying that HUB are offering different conclusions from what they stated earlier in their TS site. I would like to see the same replicated conclusion from other sites that dont contradict themselves.

The two reviews are 6 months apart. Quite possible something changed in the game.
 
Reactions: TESKATLIPOKA

amenx

Diamond Member
Dec 17, 2004
4,007
2,276
136
This is simply comical beyond belief. You...you do understand that testing the game in one area might show different results when tested in another area?
Seems all your eggs are in HUBs basket to make your point. This would be a huge issue that many sites would seek to replicate and/or find the exact same spot where it occurs. But... its only on HUB.




 

aigomorla

CPU, Cases&Cooling Mod PC Gaming Mod Elite Member
Super Moderator
Sep 28, 2005
20,877
3,228
126
Honestly, you'll probably just see it in the benchmarks. Things like a 3060 beating a 3070 Ti in FPS past a certain resolution etc.

nah.... i don't think that is possible.
Because in every benchmark the 3060ti beats a 3060 with less ram thanks to the higher core clock.

Unless the PC was using uber slow ram, like DDR3 ram, and was running off a magnetic spinner and not a SSD, i think that is probably where u'll see a bigger bottleneck before you even got the VRAM.
 

blckgrffn

Diamond Member
May 1, 2003
9,198
3,185
136
www.teamjuchems.com
nah.... i don't think that is possible.
Because in every benchmark the 3060ti beats a 3060 with less ram thanks to the higher core clock.

Unless the PC was using uber slow ram, like DDR3 ram, and was running off a magnetic spinner and not a SSD, i think that is probably where u'll see a bigger bottleneck before you even got the VRAM.

Doesn't Hogwarts already tell a different tale? And I am pretty MS Flight Sim is another that can choke out on vram levels before the core is overwhelmed.

And then we are onto looking at things like microstutters and they like that don't even show up on bar graphs because they are below the 1% minimums even. I swear some folks can feel these in their bones when it suits them and then totally not care about them later.

Clearly you can tweak settings to try and avoid this but its been this way all along. There is likely always going to be an outlier when you build your product stack and there ways that higher tiers are inferior to lower tier products.

Now whether that happens during the useful lifetime of the card appears to be a bit more subjective and coupled to the games/benchmarks you care about. You can always make up excuses why out of x, y and z that only z means anything to you.

For me, I am OK with 75-100 fps as a max, even with a 165 hz monitor. BUT I don't want really low lows, lets keep it between 75 and 100 please. That's personal preference.
 

Golgatha

Lifer
Jul 18, 2003
12,682
1,598
126
nah.... i don't think that is possible.
Because in every benchmark the 3060ti beats a 3060 with less ram thanks to the higher core clock.

Unless the PC was using uber slow ram, like DDR3 ram, and was running off a magnetic spinner and not a SSD, i think that is probably where u'll see a bigger bottleneck before you even got the VRAM.

Except....when it doesn't. The 3060 is beating the 3060 Ti and 3070 Ti due to not enough VRAM at 1440p with RT enabled. You also get weirdness like a 3090 beating a 4070 Ti by 30% at 4K with RT enabled.



"At 1440p with RT enabled we see that there are very few GPUs that can handle ray tracing in Hogwarts Legacy, at least when using ultra settings. Although the GeForce RTX 4090 was good for 85 fps and the RTX 4080 79 fps, most GPUs struggled to even hit 60 fps, such as the previous-gen flagship RTX 3090 Ti.

12GB of VRAM is the minimum here, the RTX 2080 Ti did okay but 1% lows did suffer and this was also the case with the RTX 3080. Parts like the 3070 Ti were completely broken, leading to competing parts such as the Radeon 6800 delivering twice the performance, but really it's no comparison as the 3070 Ti wasn't even remotely playable by anyone's standards whereas the Radeon 6800 was playable, not by my standards, but it was technically playable."
 

BFG10K

Lifer
Aug 14, 2000
22,709
2,979
126
Seems all your eggs are in HUBs basket to make your point. This would be a huge issue that many sites would seek to replicate and/or find the exact same spot where it occurs. But... its only on HUB.
That's not answering the question.

Is the video from HWUB a fabricated lie? Or are you saying the video doesn't actually exist and is a figment of our collective imagination?

It's a a simple 'yes' or 'no'. Answer the question.
 

amenx

Diamond Member
Dec 17, 2004
4,007
2,276
136
That's not answering the question.

Is the video from HWUB a fabricated lie? Or are you saying the video doesn't actually exist and is a figment of our collective imagination?

It's a a simple 'yes' or 'no'. Answer the question.
No I dont believe its "a lie". Just that other sites report no such issue, but rather smooth or "fluent" gameplay at the same settings.

Look I do believe vram is important, and you posted good (actually better) examples in your OP. Just not a fan of the generalized assumption that when a game allocates more vram than a card has then performance will automatically tank. Caching is often at play.

Also this thread was started quite a while back (2001) and the arguments, examples presented back then were not as compelling (to me) as the updates, examples you added to it since. Well done.

There is also a strong belief on my part that GPU manufacturers overly influence developers to make their games as difficult to run as possible and that includes going extra heavy on the vram, even when not necessary. Its the way they stay in business.
 

jpiniero

Lifer
Oct 1, 2010
14,835
5,452
136
Should point out that Hogwarts Legacy is technically a next gen console game... Yeah they are going to release it on previous gen consoles (including the Switch) but presumably they are working on an inferior version for those right now. Letting Vram go wild shouldn't be an issue for the current gen consoles.
 
Reactions: Tlh97 and Leeea

CP5670

Diamond Member
Jun 24, 2004
5,526
604
126
I went through Levitation mod for HL Alyx (excellent mod btw) and it routinely goes to around 20GB on the G2 headset. Alyx seems to use as much memory as available, and just adjusts the resolution down if there is less memory.
 
Reactions: igor_kavinski

gamervivek

Senior member
Jan 17, 2011
490
53
91
Performance drop is one effect, but most games pull down the texture quality before they show stutters, if at all. Benchmarking sites that don't play for long, would never notice even the stutters because often the game loads fine for the VRAM available, but slowly consumes more over a period.

Far Cry 6 had a hue outcry on Ubisoft forums when 3080 was having blurry textures in the game. 1.4k posts thread on it.

 

CP5670

Diamond Member
Jun 24, 2004
5,526
604
126
Performance drop is one effect, but most games pull down the texture quality before they show stutters, if at all. Benchmarking sites that don't play for long, would never notice even the stutters because often the game loads fine for the VRAM available, but slowly consumes more over a period.

Far Cry 6 had a hue outcry on Ubisoft forums when 3080 was having blurry textures in the game. 1.4k posts thread on it.


Alyx does this too. The game runs well on many different cards but drops the resolution down in different parts of the field of view, and it can be quite noticeable.
 

BFG10K

Lifer
Aug 14, 2000
22,709
2,979
126
A 16GB 4070 would make the 12GB 4070TI / 3080TI garbage in comparison, much like the 3060 did to the 3070 / 3080.

I really wonder if they'll do it. You never know given the VRAM gymnastics nVidia have been doing since Turding.

 
Jul 27, 2020
17,916
11,688
116
They just might. They know that the low VRAM is making them look pretty bad in benchmarks and now they get to charge possibly up to $100 more for the highest VRAM configuration. But that will be MSRP. Actual prices will be crazy high for the 16GB one.
 
sale-70-410-exam    | Exam-200-125-pdf    | we-sale-70-410-exam    | hot-sale-70-410-exam    | Latest-exam-700-603-Dumps    | Dumps-98-363-exams-date    | Certs-200-125-date    | Dumps-300-075-exams-date    | hot-sale-book-C8010-726-book    | Hot-Sale-200-310-Exam    | Exam-Description-200-310-dumps?    | hot-sale-book-200-125-book    | Latest-Updated-300-209-Exam    | Dumps-210-260-exams-date    | Download-200-125-Exam-PDF    | Exam-Description-300-101-dumps    | Certs-300-101-date    | Hot-Sale-300-075-Exam    | Latest-exam-200-125-Dumps    | Exam-Description-200-125-dumps    | Latest-Updated-300-075-Exam    | hot-sale-book-210-260-book    | Dumps-200-901-exams-date    | Certs-200-901-date    | Latest-exam-1Z0-062-Dumps    | Hot-Sale-1Z0-062-Exam    | Certs-CSSLP-date    | 100%-Pass-70-383-Exams    | Latest-JN0-360-real-exam-questions    | 100%-Pass-4A0-100-Real-Exam-Questions    | Dumps-300-135-exams-date    | Passed-200-105-Tech-Exams    | Latest-Updated-200-310-Exam    | Download-300-070-Exam-PDF    | Hot-Sale-JN0-360-Exam    | 100%-Pass-JN0-360-Exams    | 100%-Pass-JN0-360-Real-Exam-Questions    | Dumps-JN0-360-exams-date    | Exam-Description-1Z0-876-dumps    | Latest-exam-1Z0-876-Dumps    | Dumps-HPE0-Y53-exams-date    | 2017-Latest-HPE0-Y53-Exam    | 100%-Pass-HPE0-Y53-Real-Exam-Questions    | Pass-4A0-100-Exam    | Latest-4A0-100-Questions    | Dumps-98-365-exams-date    | 2017-Latest-98-365-Exam    | 100%-Pass-VCS-254-Exams    | 2017-Latest-VCS-273-Exam    | Dumps-200-355-exams-date    | 2017-Latest-300-320-Exam    | Pass-300-101-Exam    | 100%-Pass-300-115-Exams    |
http://www.portvapes.co.uk/    | http://www.portvapes.co.uk/    |