• Welcome to TechPowerUp Forums, Guest! Please check out our forum guidelines for info related to our community.
  • The forums have been upgraded with support for dark mode. By default it will follow the setting on your system/browser. You may override it by scrolling to the end of the page and clicking the gears icon.

GPU Test System Update May 2019

W1zzard

Administrator
Staff member
Joined
May 14, 2004
Messages
28,935 (3.75/day)
Processor Ryzen 7 5700X
Memory 48 GB
Video Card(s) RTX 4080
Storage 2x HDD RAID 1, 3x M.2 NVMe
Display(s) 30" 2560x1600 + 19" 1280x1024
Software Windows 10 64-bit
In anticipation of AMD Navi we've updated our graphics card test system using the latest Intel hardware. The game selection has been revamped, too, seeing the addition of the latest titles, and all major graphics cards are present. In total, we have 18 cards, each tested at three resolutions.

Show full review
 
Last edited by a moderator:
Looks good.

Would be nice to see one or two cards from older generations too for those who are upgrading from older hardware. Maybe a 970/980 and 7970/280x?
 
What about framerate consistency and latency, the most important aspect of a gaming card making a game "feel smooth"?
I think its long overdue.
You can always extrapolate those from the other results, can't you? When would a card with better average FPS, give you a more inconsistent experience than a slower one?
 
You can always extrapolate those from the other results, can't you? When would a card with better average FPS, give you a more inconsistent experience than a slower one?
This is exactly why media started using FPS statistics to show readers what kind of experience they should expect. An inconsistent framerate that averages higher would either make a user want the card, or not, depending on which review he reads \ watches

PS- that includes various framedrops, that can definitely effect user experience
 
Last edited:
Uhm I thought Rage2 uses only Vulkan api, is there dx12 too? Glad to see video benchmark updated, blu rays are starting to get really obsolete.
 
You can always extrapolate those from the other results, can't you? When would a card with better average FPS, give you a more inconsistent experience than a slower one?
You can't, that's the point. Frame dropping every 10 seconds will not affect average framerate by a lot, but will definitely be noticeable during gameplay.
 
I think average framerate in a demanding scene is much better than too much focusing on Lows.
1% and 0.1 lows can be just a random stutter during gameplay that can happen on any GPU.
When I'm benchmarking myself I see terrible 0.1 lows on some games and based on that you'd think the experience is terrible, while it's not.
I'm not saying 0.1/1% low results aren't valuable or shouldn't be there though; when it comes to low-end cards with limited memory sizes lows are actually useful, but other than that, rarely useful when comparing GPUs.
 
Low end cards=2080??

Even high end cards can get pushed over VRAM or other hardware limits that cause stutterring by quite a lot of games...
 
high end cards can get pushed over VRAM

It's really hard to push over the memory budget and make an 8GB card stutter related to that; that's why I said rarely useful and not useless.
 
Uhm I thought Rage2 uses only Vulkan api, is there dx12 too? Glad to see video benchmark updated, blu rays are starting to get really obsolete.
Fixing, my bad
 
Would love GN levels of testing, but that takes a lot of time, so this is a good compromise really.

But id still like to say, i would -love- for some DPC numbers when testing motherboards!
 
@W1zzard always working to get the best results and have the most complete reviews, hats off to you.
 
Yep, I will join the '99th percentile' frametimes are nice to have crowd.

Or lacking that, at least a minimum FPS number for each game, because that dóes show some interesting trends between games and cards and its not nitpicky like 0.1% frametime analysis tends to be. I also think that its more doable in terms of time investment. You already practically have the numbers.

Other than that, sweet upgrade!
 
How many games, cards, resolutions does GN test?

GN uses a smaller test selection, but they test much more rigorously, report more data, and choose their test titles with very stringent criteria - for example, you test Civilisation VI for GPU reviews. I would argue that's a pointless thing to do unless this is also your CPU test bench, since Civilisation VI is almost always CPU limited, not GPU. GN would also criticise the use of Civ VI as a benchmark in general, as they've seen patches for Civ VI impact performance dramatically in the past, making their previous test data inapplicable.

GN also reports 1% and 0.1% lows in framerate, but, beyond that, they generate frametime plots like these ones, that show in a much more significant manner, issues in specific games on specific hardware.

For example, in this test you can clearly see that the 1050Ti provided not only a lower overall framerate (longer frametimes), but it provided it less consistently overall, meaning more perceptible stutter.

123413




They test F1 2018, Apex Legends, Sniper Elite 4, Far Cry 5, GTA V, and Shadow of the Tomb Raider, at 1080p and 1440 for all games and Sniper Elite at 4K because they've found it to be a repeatable and well optimised DX12 title.

Here's a video from them outlining their own updates to their testing methodology in the first segment:
 
why remove Deus Ex but not the Witcher 3? as far as I know Deus Ex looks better and is more demanding on the GPU than the Witcher 3.
 
Interesting points

I would argue that's a pointless thing to do unless this is also your CPU test bench, since Civilisation VI is almost always CPU limited, not GPU.
Not in DX12. And a CPU limited title can be useful to detect differences in CPU load between drivers

zx71zwmkbm.jpg


as they've seen patches for Civ VI impact performance dramatically in the past, making their previous test data inapplicable.
this applies to nearly every game out there - don't reuse data

why remove Deus Ex but not the Witcher 3? as far as I know Deus Ex looks better and is more demanding on the GPU than the Witcher 3.
witcher 3 is much much more popular, and A LOT of people read its page in reviews (I checked the stats), but it'll go at some point, too
 
should we not at least wait for the next non-beta drivers from AMD?
 
Yep, I will join the '99th percentile' frametimes are nice to have crowd.

Or lacking that, at least a minimum FPS number for each game, because that dóes show some interesting trends between games and cards and its not nitpicky like 0.1% frametime analysis tends to be. I also think that its more doable in terms of time investment. You already practically have the numbers.

Other than that, sweet upgrade!

Well jay for percentile nay for minimum. Minimum does not really tell anything valuable and is very meaningless most of the time, while percentile tells how consistent frame rates game runs.
 
I thought World of Warcraft was part of the testing suite. Did it get removed previously?
 
I thought World of Warcraft was part of the testing suite. Did it get removed previously?
Yeah I think like 2 years ago. The problem was the always-online nature of the game and Blizzard pushing patches literally hours before graphics card review NDA was up. So I dropped it and only 3-4 people complained over the years
 
Hm, no waiting for Zen2. I wonder what you know that I don't...
 
Back
Top