I thought the point of the crossfire demo was to show that with the game features maxed out the AMD cards had tons of headroom left, the gtx1080 was basically running flat out. It was pretty silly since showing 1 $250 AMD card matching the gtx1080 would have been a lot more impressive.
-
Competitor rules
Please remember that any mention of competitors, hinting at competitors or offering to provide details of competitors will result in an account suspension. The full rules can be found under the 'Terms and Rules' link in the bottom right corner of your screen. Just don't mention competitors in any way, shape or form and you'll be OK.
AMD Polaris architecture – GCN 4.0
- Thread starter Boomstick777
- Start date
Or maybe the 480 can't beat 1080 but AMD doesn't have any 1070 to test against, they put 2 cards and beat the 1080 that way, saying 'our solution is cheaper and faster than competition'.
A 95% utilized P10 at 45-50 fps against 60 fps of GTX1080 for a 1/3 of the price would have been way more impressive tho. What res was it, 1440p or 4K?
A 95% utilized P10 at 45-50 fps against 60 fps of GTX1080 for a 1/3 of the price would have been way more impressive tho. What res was it, 1440p or 4K?
I think AMD have achieved exactly what they set out to do. Get all the tech forums speculating and talking about Polaris and not Pascal whilst not giving too much away.
I may even get a 480 as a stop gap until Vega. At the price it's a no brainer and lower power usage and heat to boot over my current setup.
I may even get a 480 as a stop gap until Vega. At the price it's a no brainer and lower power usage and heat to boot over my current setup.
Soldato
- Joined
- 30 Nov 2011
- Posts
- 11,493
Two 480's beat a 1080, but at what settings?
http://fudzilla.com/news/graphics/4...0-crossfire-rendering-differently-vs-gtx-1080
http://fudzilla.com/news/graphics/4...0-crossfire-rendering-differently-vs-gtx-1080
I thought everyone was assuming lower setting based on what Raja was saying about people with really good vision noticing slight differences.
Interesting they even used different versions of the game. Not saying that makes a difference just thought if you were presenting to the world you would want it all the same
Providing they used the same settings then I don't think AMD have much control over the procedural parts.
If they haven't then I guess they wanted just that little bit more fps to flaunt.
Last edited:
Soldato
- Joined
- 19 Feb 2011
- Posts
- 5,849
Wouldnt you really need to run it on a pretty much identical rig with similar drivers?
Soldato
- Joined
- 30 Nov 2011
- Posts
- 11,493
but look at the images - the 480's arent rendering the same level of detail
It would help yes but a 5930k and 2400mhz ram is a pretty common bench set up
Just as an example, Fury X scores this so its unlikely it is going to be near Fury performance. ..
http://www.tweaktown.com/articles/7...cf-vs-titan-sli-ashes-singularity/index3.html
Caporegime
- Joined
- 18 Oct 2002
- Posts
- 33,188
I'm not sure where you're getting that but the exact same game version is listed for both setups.
In terms of differences I've seen this before from multiple sites benching now since the ATOS benchmark came out, Nvidia puts in bolder colours but AMD appears to have more snow. Fud manages to say AMD has more snow but then calls it out as having less chasm detail, if Fud had ever been anywhere near snow, snow washes out detail in the same way. Koduri telling them to ask the game maker directly which is a more accurate representation of what they wanted.
The issue of it looking different on AMD and Nvidia has come up on OCUK before in similar videos/image comparisons. To me it looks like Nvidia has less detail but then has bolder colours and highlights a few things like sticking out bits of rock which gives it the appearance of more detail. It's really up to the devs, if the idea is more snow but Nvidia is tweaking a setting which reduces snow and improves performance or it could be the other way around, AMD is tweaking a setting and adding snow which increases performance. Either way, every single comparison I see the AMD and Nvidia demos I see always look different.
In terms of differences I've seen this before from multiple sites benching now since the ATOS benchmark came out, Nvidia puts in bolder colours but AMD appears to have more snow. Fud manages to say AMD has more snow but then calls it out as having less chasm detail, if Fud had ever been anywhere near snow, snow washes out detail in the same way. Koduri telling them to ask the game maker directly which is a more accurate representation of what they wanted.
The issue of it looking different on AMD and Nvidia has come up on OCUK before in similar videos/image comparisons. To me it looks like Nvidia has less detail but then has bolder colours and highlights a few things like sticking out bits of rock which gives it the appearance of more detail. It's really up to the devs, if the idea is more snow but Nvidia is tweaking a setting which reduces snow and improves performance or it could be the other way around, AMD is tweaking a setting and adding snow which increases performance. Either way, every single comparison I see the AMD and Nvidia demos I see always look different.
Huh? There's no screenshots attached to those official benchmarks (benchmark was run on the exact same settings). The ones you saw during AMD's presentation looked different because of the game's procedural weather effects: explanation
Well from the links in the thread you see one is version 1.11 one is 1.12
look at top of game configuration settings
As i said it might not make a difference i dont know just seemed odd
Caporegime
- Joined
- 18 Oct 2002
- Posts
- 33,188
Except the link you quoted they are both v1.12.19928
If you scroll down to the actual images they arent. Ignore the table as we dont seem to have actual benchmarks links for that or the table versions are mistaken
Edit
https://forums.overclockers.co.uk/showpost.php?p=29572516&postcount=526
As you can See Matt has confirmed these are in fact the links and therefore the table is wrong
Last edited:
Caporegime
- Joined
- 18 Oct 2002
- Posts
- 33,188
Neither of which were run yesterday live backstage. Those numbers were from last week, the settings used in the show and that represent those images with the difference are shown in the table.
A big hint is neither of those show the numbers shown on stage, 62.5 vs, 58.7, those links are for 1440p and both get around 50fps, the demo was 1080p and around 60fps.
From AMD MATT himself yet we have a version change.
If you have the links to the actual 1080p live version benchmarks feel free to link those. As those benchmarks matt has confirmed as run by them, are what was in the post you stated didnt have different versions.
A big hint is when its explained to you what benchmarks are in a mentioned in post then your supplied a link to the amd rep giving those same benchmarks thats most likely the benchmarks the poster is talking about.
