• 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 VEGA confirmed for 2017 H1

Status
Not open for further replies.
The Fury beats the 1060 in most situations? The 1070 I agree with, can you show me links where the 1060 makes the Fury look pathetic please? As I did consider changing to the 1060, but it looked to be a worse card to me? Im at 1440 if that makes much difference.

Hmm lets see.

1060 6GB =

Cheaper
Faster on average
2GB more VRAM
HALF the TDP (can save ££ on power supply)
Cooler
Quieter

That's why the Fury's look pathetic to me at this point in time. I'd only recommend one over a RX480/1060 6GB if you could buy a Fury dirt cheap, say £150-200 or so.

Plus Vega is around the corner, which will hopefully knock all the NVIDIA options down a pricing tier or so!
 
Hmm lets see.

1060 6GB =

Cheaper
Faster on average
2GB more VRAM
HALF the TDP (can save ££ on power supply)
Cooler
Quieter

That's why the Fury's look pathetic to me at this point in time. I'd only recommend one over a RX480/1060 6GB if you could buy a Fury dirt cheap, say £150-200 or so.

Plus Vega is around the corner, which will hopefully knock all the NVIDIA options down a pricing tier or so!

The 1060 is not faster on average.
 
Hmm lets see.

1060 6GB =

Cheaper
Faster on average
2GB more VRAM
HALF the TDP (can save ££ on power supply)
Cooler
Quieter

That's why the Fury's look pathetic to me at this point in time. I'd only recommend one over a RX480/1060 6GB if you could buy a Fury dirt cheap, say £150-200 or so.

Plus Vega is around the corner, which will hopefully knock all the NVIDIA options down a pricing tier or so!
Any chance you can link me some benchmarks which shows the 1060 faster on average? I agree it has 6gb but the fury has 4gb hbm, which equates to about 6gb normal memory. The rest are fair points, although my nitro is pretty quiet but it is massive compared to the 1060s.
 
Any chance you can link me some benchmarks which shows the 1060 faster on average? I agree it has 6gb but the fury has 4gb hbm, which equates to about 6gb normal memory. The rest are fair points, although my nitro is pretty quiet but it is massive compared to the 1060s.

HBM is the same as normal memory, on the Fury series at least. It just provides more bandwidth. Don't believe the marketing or BS from the truly die hard AMD fanboys.

I really couldn't be bothered going to look for graphs, as you should be able to google this yourself, however here we go:

873zbxy.png

Lots of the reviews only show the Fury X, which is obviously faster than the air cooled fury since it's fully unlocked and runs at higher clock speeds etc, so just deduct 8-10% from the Fury X scores for the Fury results.

You'll be able to find plenty of links where the fury is ahead of the 1060 6GB, though the fact they are so close and that the 1060 6GB actually beats the Fury is what's so pathetic IMO.
 
Hmm lets see.

1060 6GB =

Cheaper
Faster on average
2GB more VRAM
HALF the TDP (can save ££ on power supply)
Cooler
Quieter

That's why the Fury's look pathetic to me at this point in time. I'd only recommend one over a RX480/1060 6GB if you could buy a Fury dirt cheap, say £150-200 or so.

Plus Vega is around the corner, which will hopefully knock all the NVIDIA options down a pricing tier or so!

1060 is slower than FuryX and Fury in all but one DX12 game, and on the significant majority of the DX11 games.

And here is a current review with the new drivers from both companies

http://www.guru3d.com/articles_pages/gigabyte_geforce_gtx_1080_aorus_xtreme_review,12.html
 
So away from the bickering from a moment.

I've lost track. Is Vega just suppose to be the high end only (1080 and up) or is there meant to be something in the 1070 ish position as well or not known yet?

Thanks.

Vega will be a complete line up, top to bottom.

We think the high end will release first, followed by the mid and low end in the following weeks/months.

Hence the fire sales on the RX480's worldwide recently etc..

Taking questions after the initial briefing AMD CEO, Lisa Su, answered a question about rising attach rates of AMD GPUs once their Ryzen CPUs launch explaind "I think we have Ryzen launching in early March and then we'll have Vega, our enthusiast GPU launching in the second quarter."

"We should see Ryzen doing very well in the high-end," Su continued, "as well as Vega and by nature, since both of those high-end markets are markets that we don't have significant presence today, there will be an opportunity to both gain share as well as increase attach rates in those markets."


Note the repeated use of the phrases "enthusiast" and "high-end".

I think I'm going to revise my pricing expectations upward.
 
HBM is the same as normal memory, on the Fury series at least. It just provides more bandwidth. Don't believe the marketing or BS from the truly die hard AMD fanboys.

I really couldn't be bothered going to look for graphs, as you should be able to google this yourself, however here we go:

873zbxy.png

Lots of the reviews only show the Fury X, which is obviously faster than the air cooled fury since it's fully unlocked and runs at higher clock speeds etc, so just deduct 8-10% from the Fury X scores for the Fury results.

You'll be able to find plenty of links where the fury is ahead of the 1060 6GB, though the fact they are so close and that the 1060 6GB actually beats the Fury is what's so pathetic IMO.

finding one or two abnormal games where furyx sux to set it as a reference doesn't make your point.
and using that argument to discredit HBM bandwidth doesn't work either, why dont you compare 1440p and 4k results to disprove the merite of HBM, unless you consider 980ti performance is on par with 1060...
 
1060 is slower than FuryX and Fury in all but one DX12 game, and on the significant majority of the DX11 games.

And here is a current review with the new drivers from both companies

http://www.guru3d.com/articles_pages/gigabyte_geforce_gtx_1080_aorus_xtreme_review,12.html

I see Guru3D are still only benchmarking Resident Evil 7 using settings that stops AMD cards from utterly destroying nVidia's cards.

If your an 8GB AMD user, like an RX 480 or 390X turn Shadow Cache on, it will boost performance by about 40%


HBM is the same as normal memory, on the Fury series at least. It just provides more bandwidth. Don't believe the marketing or BS from the truly die hard AMD fanboys.

I really couldn't be bothered going to look for graphs, as you should be able to google this yourself, however here we go:

873zbxy.png
Lots of the reviews only show the Fury X, which is obviously faster than the air cooled fury since it's fully unlocked and runs at higher clock speeds etc, so just deduct 8-10% from the Fury X scores for the Fury results.

You'll be able to find plenty of links where the fury is ahead of the 1060 6GB, though the fact they are so close and that the 1060 6GB actually beats the Fury is what's so pathetic IMO.

How old is this review? the latest benchmarks have the RX 470 ahead of the GTX 1060 with the Fury-X faster than a GTX 1070.

untitled_1.png
 
Last edited:
Vega will be a complete line up, top to bottom.

We think the high end will release first, followed by the mid and low end in the following weeks/months.

Hence the fire sales on the RX480's worldwide recently etc..

Thanks

Taking questions after the initial briefing AMD CEO, Lisa Su, answered a question about rising attach rates of AMD GPUs once their Ryzen CPUs launch explaind "I think we have Ryzen launching in early March and then we'll have Vega, our enthusiast GPU launching in the second quarter."

"We should see Ryzen doing very well in the high-end," Su continued, "as well as Vega and by nature, since both of those high-end markets are markets that we don't have significant presence today, there will be an opportunity to both gain share as well as increase attach rates in those markets."


Note the repeated use of the phrases "enthusiast" and "high-end".

I think I'm going to revise my pricing expectations upward.


I just want something worthwhile from my 290. So I guess "enthusiast and high end" is where I would be looking anyway. Just hopefully that's not just 1080+ priced cards only.

Oh well, been putting some away each month anyway, so hopefully have enough for something good, still cry when handing it over though :D
 
HBM is the same as normal memory, on the Fury series at least. It just provides more bandwidth. Don't believe the marketing or BS from the truly die hard AMD fanboys.

I really couldn't be bothered going to look for graphs, as you should be able to google this yourself, however here we go:

873zbxy.png

Lots of the reviews only show the Fury X, which is obviously faster than the air cooled fury since it's fully unlocked and runs at higher clock speeds etc, so just deduct 8-10% from the Fury X scores for the Fury results.

You'll be able to find plenty of links where the fury is ahead of the 1060 6GB, though the fact they are so close and that the 1060 6GB actually beats the Fury is what's so pathetic IMO.

You do realise that you're both comparing a 28nm couple year old card on a 4 year old architecture to something that just came out on 16nm.

I don't know if you know how to read graphs, but a stock 1060 gets 64fps average, the overclocked G1 gaming 1060 gets 71fps average and the Fury X gets 80FPS average. The minimum is lower.... but the average on a stock Fury X is a full 25% faster than the 1060 stock.... and you're using that to imply the 1060 is faster in this benchmark, it's not.

The Fury X min is higher than the stock 1060 but slightly lower than the overclocked 1060, however if the minimums were happening frequently, the average would be lower. Minimums on their own can be useful, but can also be worthless, if it's one frame in 5000, often at the start of a benchmark during loading, then it means nothing, if it happens constantly, it matters. Average FPS is still fairly meaningful but again take into account to how many frames are in/around the minimum.

That graph, the one you found that apparently shows the 1060 faster, actually has the Fury X a full 25% faster and a Fury would still be 15% faster than a 1060.

Also at current prices there have been some epic deals on Fury X's which run cooler and quieter than a 1060 for with the deals in the £250 range, around the same price. Today I'd buy a Fury X in particular but probably a Fury also over a 1060. Neither is going to be good enough for 4k gaming(in general on newer graphically demanding games). There are still incredibly few games that genuinely need more than 4GB and most of those offer exceptionally minimal texture resolution difference. Without ultra high textures(with marginal or literally no improvement at all, with literally not one game where highest qual textures makes any measurable IQ difference), no games need over 4GB of graphics memory yet. Hell factor in freesync vs gsync, and your sale price Fury X + freesync will still come in WAY under a 1060 + g-sync monitor.

Saving £££'s on a PSU... in what world? A Fury X will run overclocked in a heavily overclocked CPU system entirely fine on a 500W PSU. Those who buy 700-1200W PSU's for a single card are just wasting money. Due to how PSU's are divided up into segments there is really no cheap PSU options that will save you anything over a PSU that can run a Fury X anyway.
 
Last edited:
Hmm lets see.

1060 6GB =

Cheaper
Faster on average
2GB more VRAM
HALF the TDP (can save ££ on power supply)
Cooler
Quieter

That's why the Fury's look pathetic to me at this point in time. I'd only recommend one over a RX480/1060 6GB if you could buy a Fury dirt cheap, say £150-200 or so.

Plus Vega is around the corner, which will hopefully knock all the NVIDIA options down a pricing tier or so!

The 1060 is not faster on average then a Fury. In fact the 1060 is slower than a gtx980 but the Fury is quite a bit faster. I don't know why you think that tbh.
 
Plus as we know Fury performance tends to be resolution dependent.

AMD better hope that Vega isn't!!!
Very few people are going to be spending £400+ on a GPU to run at 1080p. Vega needs to perform at 1440p and 4k.

If they really are doing a whole stack of SKUs on Vega the the lower cards will have need good 1080p performance, but they will be adjusted for that accordingly.
 
Right now Fury could be had a bargain, not much more expensive that RX480 but performance wise firmly ahead.
 
Furies are going for £275 cheapest, at best, but closer to £300. You can get a beast 480 (Devil or MSI) for £220. So 10% more performance & half the vram for £50-75 more and the longest ass card in existence? No thanks.
 
Furies are going for £275 cheapest, at best, but closer to £300. You can get a beast 480 (Devil or MSI) for £220. So 10% more performance & half the vram for £50-75 more and the longest ass card in existence? No thanks.

Not long ago the 480's were going at the same price as those Furies in many cases. I got my Fury X last year after I cancelled my order for a 480 Devil, only cost me £50 more.

If the 480's launched at these prices, which are their original stated pricing the Fury's wouldn't have sold as well as they did.
 
Furies are going for £275 cheapest, at best, but closer to £300. You can get a beast 480 (Devil or MSI) for £220. So 10% more performance & half the vram for £50-75 more and the longest ass card in existence? No thanks.

Is not simple cut as you describe it. At resolutions 2560x1440 and above, the GDDR5 memory is worst than the HBM and the GDDR5X regardless how much you have as long as you do not run out of it.
And given that 4GB HBM is more than adequate for all but 2 games.

Hence the Fury's seal club even the GTX1070 at higher resolutions, having "less" VRAM. And we are talking about stock Fury's.
Even the worst one can take 10% VRAM overclock, which makes a HUGE difference due to bigger bandwidth.
 
Is not simple cut as you describe it. At resolutions 2560x1440 and above, the GDDR5 memory is worst than the HBM and the GDDR5X regardless how much you have as long as you do not run out of it.
And given that 4GB HBM is more than adequate for all but 2 games.

Hence the Fury's seal club even the GTX1070 at higher resolutions, having "less" VRAM. And we are talking about stock Fury's.
Even the worst one can take 10% VRAM overclock, which makes a HUGE difference due to bigger bandwidth.

You are just complicating things. Common sense dictates to go with the higher amount of VRAM if you are a 1440P gamer and it is cheaper. Even 1080P could warrant the bigger amounts now also and there is more than 2 games that require more than 4 GB for 1440P.
 
You are just complicating things. Common sense dictates to go with the higher amount of VRAM if you are a 1440P gamer and it is cheaper. Even 1080P could warrant the bigger amounts now also and there is more than 2 games that require more than 4 GB for 1440P.

Actually the whole subject is complicated because we should take into consideration a lot of parameters, is not a simple matter how much VRAM exists.

The speed of the VRAM is a major factor also, when comes to games that do not use all the VRAM.

Also don't forget that AMD on their HBM cards support texture compression, and I have seen it on games like TW Warhammer, Stellaris, XCOM 2 when tested the GTX1080 against the Nano.

The Nano is using 2/3 of the VRAM the 1080 did on those games.
Especially XCOM 2, was consuming 3.8GB on the Nano and 5.2 on the 1080.
Same settings at 2560x1440 on all games.
 
Status
Not open for further replies.
Back
Top Bottom