DavidGraham
Veteran
NVIDIA GPUs recieved a boost from the new driver as well.
But once more, this video features a highly overclocked 7900XTX, results are not applicable here, Stick to stock vs stock.
IHV Business strategies and consumer choice
- Thread starter eastmen
- Start date
NVIDIA GPUs recieved a boost from the new driver as well.
But once more, this video features a highly overclocked 7900XTX, results are not applicable here, Stick to stock vs stock.
A stock 7900 XTX likely still demolishes the 4080 in AC Mirage by a ~20% margin and your graph uses a 4090 to compare the different driver revisions so there's no actual data for the 4080 ...
DavidGraham
Veteran
No data vs the 4080 either, so that's a pure speculation on your part. We should talk again only when you have stock vs stock data.
The 4080 was MAYBE only able to get up to par with the 7900 XTX after it's driver update before the latter is rolled out with it's first October driver. I don't need to extrapolate as hard as you do to conclude that the 4080 is still getting utterly massacred either way in AC Mirage ...
D
Deleted member 2197
Guest
Not updated with latest Nvidia driver but includes AMD latest driver.
Assassin's Creed Mirage: PC Performance Benchmarks (updated with Adrenalin Edition 23.20.11.7 drivers) | RPG/Role Playing | GPU TEST
We tested Assassin's Creed Mirage at the highest graphics settings, on video cards from the GEFORCE RTX and RADEON RX series. Also, during the tests we conducted, we assessed the performance and qu
gamegpu.tech
AMD's latest driver is 23.10.1 which proceeds after version 23.20.11.7 tested in the link. It was specifically version 23.10.1 that he saw a massive ~20% performance boost spike at 4K Ultra settings on the 7900 XTX. Meanwhile the graph from DSO only showed a 6.5% improvement with the same settings on a 4090 ...Not updated with latest Nvidia driver but includes AMD latest driver.
Assassin's Creed Mirage: PC Performance Benchmarks (updated with Adrenalin Edition 23.20.11.7 drivers) | RPG/Role Playing | GPU TEST
We tested Assassin's Creed Mirage at the highest graphics settings, on video cards from the GEFORCE RTX and RADEON RX series. Also, during the tests we conducted, we assessed the performance and qugamegpu.tech
Before 23.10.1, the 7900 XTX was slightly ahead of the 4080 in the in-game benchmark. Even after factoring both the different SKU and after driver updates, the 7900 XTX is unquestionably now a tier above the 4080 in AC Mirage ...
D
Deleted member 2197
Guest
Percent improvement will normally vary by benchmarked scene and review site. Since the benchmarks I posted are dated it's likely best as @DavidGraham mentioned to wait for reliable benchmarks using the latest drivers.
Nearly everyone used the automated in-game benchmark for AC Mirage including gameGPU so we can factor out the "scene differences" between most reviews sites so it makes doing comparisons much easier with different setups. That guy in the video went from 96FPS (which nearly checks out with PCGH's results) to 117FPS which is a ~20% upswing ...
D
Deleted member 2197
Guest
The 23.10.1 is currently not available for download on AMD's site. Guess some review sites are waiting for the next official download.
DavidGraham
Veteran
Which is barely enough to make the 7900XTX equal to the 4080, considering -at launch- the 4080 was 20% faster at 1440p and 15% faster at 2160p. (just like I reported back then)
Assassin's Creed Mirage GPU Benchmarks: NVIDIA RTX 40 vs. AMD RX 7000 | Hardware Times
Assassin’s Creed: Mirage is out across all major platforms, giving gamers a glimpse into 13th-century Baghdad, the heart of the Abbasid Caliphate. We tested the Anvil based-title using the latest NVIDIA and AMD graphics cards across 1080p, 1440p, and 4K, and have compiled the results below...
www.hardwaretimes.com
Anyway, here is the most direct head to head comparison, with the latest drivers. They are equal to each other.
Last edited:
D
Deleted member 2197
Guest
Is that normal wattage usage for the 7900 XTX? Huge difference compared to the RTX 4080.
Doesn't really matter since anyone can install the driver ...
Back to cherrypicking I see and with results out of line in comparison to the majority as well. Nobody can't help you with five stages if you want to deny the data so badly ...
So the 4080 has to lower the resolution to come close to the 7900 XTX ?
DavidGraham
Veteran
No cherrypicking here, you have no data whatsoever to backup your claim. In the beginning 3 outlets have already shown the NVIDIA side winning by a huge margin at launch: GameGPU, DSO Gaming and Hardware Times, tests done later after the release of newer AMD drivers show AMD catching up. They should be about equal now, not the ridiculous 20% to 30% AMD advantage that you claim out of thin air with no data from direct head to head comparisons.
No, the test only shows 1440p results, 4K is probably the same or close.
GameGPU corrected their results to be inline with the others and DSO doesn't have a 4080 to test against which just leaves you with Hardware Times so thus you cherrypick their results ...
That statement is inconsistent with DSO's driver update comparison graph where the 4090 showed as 9% gain at 1440p as opposed to a 6.5 gain at 4K. Beyond this point it's easier for you to concede ...
DavidGraham
Veteran
Which others? Do you need me to list more?
More benchmarks.
늘 익숙한 맛의 유비겜, 어쌔신 크리드 미라지, 그래픽카드 벤치마크
퀘이사존 어쌔신 크리드 미라지 그래픽카드 벤치마크 과학과 신화, 역사가 어우러져 매력적인 세계관으로 유…
quasarzone.com
Assassin's Creed Mirage Performance Benchmark Review - 30 GPUs Tested
Assassin's Creed Mirage delivers a fresh experience by stripping away RPG elements, allowing players to fully engage in the series' core stealth gameplay. In our performance review, we're taking a closer look at image quality, VRAM usage, and performance on a wide selection of modern graphics cards.
www.techpowerup.com
TPU admitted to not using the in-game benchmark and you showed me some strange Korean site with very few details (especially as to what scene they tested in question) with highly suspicious results at lower resolutions which would lead anyone to conclude that their methodology is likely bunk ...
Are you going to capitulate yet ? Because I'm not interested in continuing the discussion any further ...
DavidGraham
Veteran
Meanwhile you've shown what exactly? No data, no tests, just one video of an OC'ed 7900XTX! You are discarding sites left and right, ignoring any test that isn't in-game benchmark, and then you claim I am the one doing cherrypicking!
Even the GameGPU results (with updated AMD drivers but without updated NV drivers) shows the 7900XTX is equal to the 4080 at 4K. So I am not sure what you are going on about.
Assassin's Creed Mirage: PC Performance Benchmarks (updated with Adrenalin Edition 23.20.11.7 drivers) | RPG/Role Playing | GPU TEST
We tested Assassin's Creed Mirage at the highest graphics settings, on video cards from the GEFORCE RTX and RADEON RX series. Also, during the tests we conducted, we assessed the performance and qu
gamegpu.tech
Last edited:
You are cherrypicking and after looking closer at Hardware Times results, it is strangely similar to that Korean site so I guess your back to having no reliable data. I didn't pick the in-game benchmark scene as the level playing field since Ubisoft was courteous enough to do that for us so if you somehow think that it's not representative with the game's performance then go take that issue up with Ubisoft instead ...
DavidGraham
Veteran
Thanks but no. Here are all the results of 4080 vs 7900XTX at 4K on the whole web.
GameGPU: 7900XTX = 4080
PurePC: 4080 is 2% faster
TechPowerUp: 4080 is 5% faster
HardwareTimes: 4080 is 15% faster
QuasarZone: 4080 is 17% faster
And lastly, this one with the latest drivers showing 7900XTX = 4080 @1440p, pick your poison. If you have more reliable data than this, then present them, if not give it a rest.
That's a completely inappropriate response. If the data's no good, explain why and present data that is good. Everyone is going to present data that supports their view - no-one's going around looking for data that disproves themselves. If you're lucky, you'll get someone who's more interested in the truth than an outcome and they'll present data just to add to the knowledge pool. In a public forum, these data come together from opposing viewpoints to present an overview that points to the truth. If someone's position is right, the more data we get, the more they are proven right, so no-one should be against More Data unless they fear they are wrong and they have trouble dealing with not being right 100% of the time.
In the subsequent discussion, I see DavidGraham is presenting data from numerous vendors. You've presented only one piece which has been pointed out as skewed by clockspeed. The request stands for standardised base-hardware comparison and you need to present info on this. Fixating on an in-game benchmark also isn't ideal as people don't play the benchmark. Broad sampling of the entire game across multiple users and different playthroughs will give a bit insight into how the game plays across hardware.
These kind of discussion are so dumb. They are very easy to conclude. Over time, lots of vendors do lots of tests. We just need collate all the data in one place and examine it and we'll see the truth. Someone could pony up the money and have these guys produce a meta-analysis, but we should be able to take a dozen benchmarks and get some rough averages ourselves.
