Hard OCP Blind test of Vega vs 1080Ti

[VanayadGaming]

 

 

The guys at HardOCP did a blind test with the help of AMD where they built 2 identical systems one with a 1080Ti and one with an RX Vega. Both had either freesync/gsync monitors. The thing here is that Kyle supplied his own 1080ti and did a full system reinstall just to make sure the nvidia system is not being tampered with. 

The conclusion was this:
 

Quote

Six out of the ten participants said the experience on both systems was equal, one preferred the NVIDIA GTX 1080 Ti system and three preferred the AMD RX Vega system.

 

Also, only a few said that 300$ more was worth it for one of the cards.

Well, this was tested in doom, so AMD does have an advantage there...but it looks promising indeed. I might just buy one if the prices are decent (+freesync ofc)

[MorganO]

1 minute ago, VanayadGaming said:

-snip-

i heard the sapphire cards are going to be $700-900 (air cooled to liquid)

https://www.eteknix.com/700-sapphire-radeon-rx-vega/

[VanayadGaming]

1 minute ago, KOMTechAndGaming said:

i heard the sapphire cards are going to be $700-900 (air cooled to liquid)

https://www.eteknix.com/700-sapphire-radeon-rx-vega/

expensive... but at the same time, I'm interested in the overall price of the freesync+gpu. Also I don't trust those pricing rumours.

[MorganO]

Just now, VanayadGaming said:

expensive... but at the same time, I'm interested in the overall price of the freesync+gpu. Also I don't trust those pricing rumours.

neither do i but assuming they are mostly correct then amd fudged this launch up, speciall with nvidia volta or what ever the next line up of cards that are coming

[Shimejii]

If they charge more then 500$ for this card its going to be a complete failure lol

[Misanthrope]

Hey if AMD is truly going with "Buy Vega because you can still save 300 bucks on your G-Sync monitor" then RX Vega will be fucking stillborn.

 

It's a decent argument don't get me wrong but most people are not gonna care: If you can afford 700 for a GPU you are almost always the kind of guy that also can afford a higher end monitor to go with and those are mostly similarly priced: the G-Sync costs gets eaten up fairly quickly on higher end panels.

[VanayadGaming]

Well, at the moment the difference between some freesync/gsync high end monitors is just the price... nothing else.

[tom_w141]

Blind testing is bad news. When a company blind tests instead of revealing impressive numbers it means the product is inferior. They are basically saying its not as good but hey it feels the same! This is because at higher frame rates for example, we can't notice the difference between 120 and 150 fps.

 

Also the fact that they keep stressing vega + freesync is cheaper than NVidia +gsync just to me says they know they aren't competitive on price for the gpus and are relying on the large premium carried by gsync panels. Well that's crap logic because for a start someone upgrading their gpu might already have a decent freesync monitor and therefore it would be better for them to get a 1080Ti and just not use the freesync functionality,

[Misanthrope]

1 minute ago, VanayadGaming said:

Well, at the moment the difference between some freesync/gsync high end monitors is just the price... nothing else.

Lower end, absolutely. Just to show you how much a bit of a personal case: I might but a Freesync panel next week: it's around 120 bucks 1080p 75hz freesync which is a panel that's coming out of several models including LG and Acers from what I've seen.

 

The point is that I have a 1070 gpu and I don't think I'll switch to Vega just to activate Freesync: it's just that the panels that are low priced but still have some nice features like above 60hz, IPS, etc. Just happen to have Freesync because why the hell not? It's just part of the spec.

 

Now if I wanted to spend a lot more on a screen however (which I am not, I spend enough on a platform change a week ago and now want to save for a car) then there's really not that much of a difference between say monitors on the ROG Swift it doesn't matters that much. In fact as price goes up it matters less and less.

 

So for AMD to put Freesync as a major selling point of RX Vega seems very desperate: Freesync strength is on the Polaris range of cards not here.

[LooneyJuice]

Oh not Vulkan again... Poor Doom has turned into another Ashes of the Benchmark for AMD. Even though it is an actual game, and an exceptional one at that. Not to mention, I'm not a fan of Blind Tests at all. IMO, it removes all the objectivity of painstakingly produced numbers from outlets such as GN and boils it down to a few soundbites. 

 

I could probably tailor certain games to run equally well between a 1070 and a 970 with no glaringly noticeable detriment to visuals at 1080p and have people praise one or the other.

 

Also, seriously, if it's still dodgy to base your purchase on DX12 implementation, how prudent is it to base it off Vulkan?

[Misanthrope]

5 minutes ago, tom_w141 said:

Blind testing is bad news. When a company blind tests instead of revealing impressive numbers it means the product is inferior. They are basically saying its not as good but hey it feels the same! This is because at higher frame rates for example, we can't notice the difference between 120 and 150 fps.

 

Also the fact that they keep stressing vega + freesync is cheaper than NVidia +gsync just to me says they know they aren't competitive on price for the gpus and are relying on the large premium carried by gsync panels. Well that's crap logic because for a start someone upgrading their gpu might already have a decent freesync monitor and therefore it would be better for them to get a 1080Ti and just not use the freesync functionality,

One guy actually did called it specifically and said the Nvidia system felt noticeable smoother far above 120hz and the AMD one had dips.

[tom_w141]

Just now, Misanthrope said:

One guy actually did called it specifically and said the Nvidia system felt noticeable smoother far above 120hz and the AMD one had dips.

Can't actually watch the video at work. Did they have an fps counter on screen?

[cluelessgenius]

yeah im also having a hard time believing they gonna put out a 400W 1000$ card that performs worse than a 1080. all the rumors so far seem to be negative. i mean sure theres gonna be some bad news but all of it? idk we`ll see

[Misanthrope]

4 minutes ago, tom_w141 said:

Can't actually watch the video at work. Did they have an fps counter on screen?

They didn't say but it doesn't appears to since none of them are calling specific numbers but that one guy who picked up on it.

 

They were far more interested in telling us the "gaming" credentials of the guys participating, which of course are fairly meaningless: I believe Linus is quite a fucking crap gamer and would lose any and all tournaments and such but still could call G-Sync from Freesync and what was the max refresh rate just from sight because you know, he's a reviewer and has looked at it objectively instead of a silly fucking Pepsi challenge.

[VanayadGaming]

1 minute ago, Misanthrope said:

They didn't say but it doesn't appears to since none of them are calling specific numbers but that one guy who picked up on it.

 

They were far more interested in telling us the "gaming" credentials of the guys participating, which of course are fairly meaningless: I believe Linus is quite a fucking crap gamer and would lose any and all tournaments and such but still could call G-Sync from Freesync and what was the max refresh rate just from sight because you know, he's a reviewer and has looked at it objectively instead of a silly fucking Pepsi challenge.

To be fair, another 3 said the AMD systems was better. 

Regardless, most likely the vega is slower than a 1080ti. Otherwise they would have done something similar to ryzen

[Misanthrope]

5 minutes ago, cluelessgenius said:

yeah im also having a hard time believing they gonna put out a 400W 1000$ card that perform worse than a 1080. all the rumors so far seem to be negative. i mean sure theres gonna be some be news but all of it? idk we`ll see

This was an AMD sponsored video so as bad as it sounds, no actually RX Vega looks to be as bad if AMD paid for this video to try and use a terrible argument to sell you an expensive GPU by claiming it saves money on a monitor on and almost complete non-sequitur. 

[tom_w141]

6 minutes ago, cluelessgenius said:

i mean sure theres gonna be some bad news but all of it?

Can you think of anything positive to say about a new product that competes with 1 that's been out for a year (1080) and costs more than the existing product? I can't. The poor people that have forever been saying "wait for vega" may aswell have bought a 1080 last year it would perform the same and cost them less.

 

2 minutes ago, Misanthrope said:

he's a reviewer and has looked at it objectively instead of a silly fucking Pepsi challenge.

Yeah as soon as I see blind tests and doing anything to avoid the numbers all I'm thinking is: Product DOA. Do you see Nvidia blind testing 1080Tis? NOPE. Just hand out reviewer samples and let the dominance speak for itself.

[mr moose]

I'm all in favor of double blind tests, but this is not a blind test, this is marketing.

 

I'd hate to think how big the margin of error is with a sample size of only 10. 

 

[Misanthrope]

Just now, VanayadGaming said:

To be fair, another 3 said the AMD systems was better. 

Regardless, most likely the vega is slower than a 1080ti. Otherwise they would have done something similar to ryzen

Yeah and Wendell from Level1Techs for example will fellate Vega no matter what on the strength of his irrational, Linux-fanboy driven hate for Nvidia alone, doesn't really means that AMD is clawing back GPU market share with such a lame argument, not in the high end anyway.

 

I feel that this will be an expensive mistake and that they'll focus on midrange only architecture in the future.

[tom_w141]

2 minutes ago, VanayadGaming said:

something similar to ryzen

Exactly. Testing side by side and comparing numbers side by side. That's the difference between knowing you have something good and trying to cover up a failed product.

[Taf the Ghost]

1 minute ago, Misanthrope said:

This was an AMD sponsored video so as bad as it sounds, no actually RX Vega looks to be as bad if AMD paid for this video to try and use a terrible argument to sell you an expensive GPU by claiming it saves money on a monitor on and almost complete non-sequitur. 

The weird thing is that if AMD has a GPU that slots between a 1080 & 1080 Ti, they could retail it at $599 USD and it's hard to complain about that, too much.

 

It's going to run hot/higher TDP, but at that class of buyer, it's little more than a talking point. This marketing roll out is either going to be a lesson in Anti-Hype or someone screwed up the pricing.

[Misanthrope]

Just now, Taf the Ghost said:

The weird thing is that if AMD has a GPU that slots between a 1080 & 1080 Ti, they could retail it at $599 USD and it's hard to complain about that, too much.

 

It's going to run hot/higher TDP, but at that class of buyer, it's little more than a talking point. This marketing roll out is either going to be a lesson in Anti-Hype or someone screwed up the pricing.

It looks like their gamble with HBM2 is as premature as Fiji was with HBM and the increased prices and difficulties keep raising prices on their high end stuff. I hope it pays up for them in other areas like really fancy Ryzen + Vega APUs for example but the tech is clearly not ready for prime time and I suspect the main culprit behind the really high rumored price tags.

[mr moose]

2 minutes ago, Taf the Ghost said:

The weird thing is that if AMD has a GPU that slots between a 1080 & 1080 Ti, they could retail it at $599 USD and it's hard to complain about that, too much.

 

It's going to run hot/higher TDP, but at that class of buyer, it's little more than a talking point. This marketing roll out is either going to be a lesson in Anti-Hype or someone screwed up the pricing.

I never buy the top end card, so if vega performs like a 1080 but is actually priced closer to a card I can afford, then I am much more likely to consider it. 

[Taf the Ghost]

2 minutes ago, tom_w141 said:

Can you think of anything positive to say about a new product that competes with 1 that's been out for a year (1080) and costs more than the existing product? I can't. The poor people that have forever been saying "wait for vega" may aswell have bought a 1080 last year it would perform the same and cost them less.

 

Yeah as soon as I see blind tests and doing anything to avoid the numbers all I'm thinking is: Product DOA. Do you see Nvidia blind testing 1080Tis? NOPE. Just hand out reviewer samples and let the dominance speak for itself.

Well, Nvidia pretty much just rolled out the real version of the Titan X (first one) with the 1080 Ti. So, technically Nvidia was even dogging everyone on that.

 

I saw someone mention something that could actually be the "party piece" here. We assume a linear-ish relationship between 1080p, 1440p & 4K gaming. What if some of the optimizations makes 4K run at a lot lower differential compared to 1080p? Considering the way the Fury X lined up against the 1080 on launch (did much better comparatively at 4K than a 1080p), we could be in for something wacky.

 

Because if it's a better 4K card, is that worth more than the 1080 Ti? 

 

Mostly spitballing, as I really don't know what they're up to.

[RagnarokDel]

Come on, 5X years old peeps are not the proper group to judge that kind of shit.

 

Quote

costs more than the existing product? I can't. The poor people that have forever been saying "wait for vega" may aswell have bought a 1080 last year it would perform the same and cost them less.

You have no fucking clue of how much it's going to cost. It's a fucking rumor for one country. By that logic, a 7700k costs $435, that's the price I get when I look at newegg.ca