Poll: The Vega Review Thread.

[shankly1985]

Wonder if he's still on his "I won't recommend Radeon"

People saying it's just the reporting bug that Gamers nexus reported. But Jay saying he using a driver that supposed to have fixed that.
More testing needed.

 

[TonyTurbo78]

y

Ouch at those 56 prices. The deal I got on the 64 is looking good

yep, £449 for 64 is looking ok at the moment.

 

[RavenXXX2]

 

[Gerard]

Jay two cent just got 1980 on water Vega

https://twitter.com/JayzTwoCents/status/902276773009793024

Removed it as he's unsure if it was clocking to that or not 0_o

 

[shankly1985]

1080 Good 13/10 FPS faster.. I know he likes to keep things out of the box but I do really wish he made the core clock more stable I just feel this is hurting performance. All he needs to do is drop the voltage and aim for 1600.. The Ref cooler is trash, and AMD just keep shooting themselves in the foot with this cooler.. It does nothing to help them.

 

[Scougar]

Wonder if he's still on his "I won't recommend Radeon"

It was still a lame move by AMD.

 

[Rroff]

1080 Good 13/10 FPS faster.. I know he likes to keep things out of the box but I do really wish he made the core clock more stable I just feel this is hurting performance. All he needs to do is drop the voltage and aim for 1600.. The Ref cooler is trash, and AMD just keep shooting themselves in the foot with this cooler.. It does nothing to help them.

But if you start tweaking like that you can get more out the 1080 as well.

 

[triss]

1080 Good 13/10 FPS faster.. I know he likes to keep things out of the box but I do really wish he made the core clock more stable I just feel this is hurting performance. All he needs to do is drop the voltage and aim for 1600.. The Ref cooler is trash, and AMD just keep shooting themselves in the foot with this cooler.. It does nothing to help them.

Cant really change volts etc in one card for a stock vs stock comparison

 

[AthlonXP1800]

https://twitter.com/ryanshrout/status/902252203674669057

AMD shipped so very few Vega 56 worldwide on launch day compared to Fury Pro on launch day back on 10 July 2015 saw AMD shipped 10,000 Fury Pro cards for US market alone that was very plenty of supply lasted for about a month.

RTG learnt absolutely nothing from Fiji for 2 years.

Look like Vega 56 Hynix HBM2 chips yield was far worsen than Fury Pro HBM1 chips.

 

[megatron]

Plan to get a freesync 2 monitor, on 60Hz ATM. I voted I would buy (yesterday I did).

 

[Martini1991]

Removed it as he's unsure if it was clocking to that or not 0_o

I've only started to hear about this youtuber.

I don't have much faith in him.

 

[Brazo]

1080 Good 13/10 FPS faster.. I know he likes to keep things out of the box but I do really wish he made the core clock more stable I just feel this is hurting performance. All he needs to do is drop the voltage and aim for 1600.. The Ref cooler is trash, and AMD just keep shooting themselves in the foot with this cooler.. It does nothing to help them.

A better comparison would be to overclock the 1080 to show the maximum it's easily capable of by a typical end user and then it's really game over for Vega.

 

[ICDP]

A better comparison would be to overclock the 1080 to show the maximum it's easily capable of by a typical end user and then it's really game over for Vega.

What an utterly ironic post, in response to another post bemoaning the fact the reviewer never gave the AMD card special attention. Remember the RX 64 can overclock as well. In fact it's easier to overclock an AMD card because all the overclocking tools you need are in the actual drivers. No need to install 3rd party apps.

Regardless it is of course possible to show a number of scenarios showing RX 64 soundly beating the GTX 1080. Dirt 4 for example. Or maybe the ones below from Techpowerup?

https://www.techpowerup.com/reviews/AMD/Radeon_RX_Vega_64/13.html
https://www.techpowerup.com/reviews/AMD/Radeon_RX_Vega_64/9.html
https://www.techpowerup.com/reviews/AMD/Radeon_RX_Vega_64/15.html
https://www.techpowerup.com/reviews/AMD/Radeon_RX_Vega_64/11.html
https://www.techpowerup.com/reviews/AMD/Radeon_RX_Vega_64/16.html

Or let's show the 1080 in a far better light (from the same review)
https://www.techpowerup.com/reviews/AMD/Radeon_RX_Vega_64/10.html
https://www.techpowerup.com/reviews/AMD/Radeon_RX_Vega_64/14.html
https://www.techpowerup.com/reviews/AMD/Radeon_RX_Vega_64/17.html
https://www.techpowerup.com/reviews/AMD/Radeon_RX_Vega_64/18.html

The point is, both cards can overclock and be tweaked to run better than stock. Both cards trade blows, So claiming one benchmark as an indicator is disingenuous at best and outright bias at worst.

 

[WeSbO]

https://twitter.com/ryanshrout/status/902252203674669057

AMD shipped so very few Vega 56 worldwide on launch day compared to Fury Pro on launch day back on 10 July 2015 saw AMD shipped 10,000 Fury Pro cards for US market alone that was very plenty of supply lasted for about a month.

RTG learnt absolutely nothing from Fiji for 2 years.

Look like Vega 56 Hynix HBM2 chips yield was far worsen than Fury Pro HBM1 chips.

I honestly think his source is fibbing or misunderstood something... That or something went terribly wrong, Gibbo said he got 1000s (that's just one UK e-tailer) just that he didn't receive all of his stock, but not because of AMD but because of the bank holiday.

 

[Brazo]

What an utterly ironic post, in response to another post bemoaning the fact the reviewer never gave the AMD card special attention. Remember the RX 64 can overclock as well. In fact it's easier to overclock an AMD card because all the overclocking tools you need are in the actual drivers. No need to install 3rd party apps.

Regardless it is of course possible to show a number of scenarios showing RX 64 soundly beating the GTX 1080. Dirt 4 for example. Or maybe the ones below from Techpowerup?

https://www.techpowerup.com/reviews/AMD/Radeon_RX_Vega_64/13.html
https://www.techpowerup.com/reviews/AMD/Radeon_RX_Vega_64/9.html
https://www.techpowerup.com/reviews/AMD/Radeon_RX_Vega_64/15.html
https://www.techpowerup.com/reviews/AMD/Radeon_RX_Vega_64/11.html
https://www.techpowerup.com/reviews/AMD/Radeon_RX_Vega_64/16.html

Or let's show the 1080 in a far better light (from the same review)
https://www.techpowerup.com/reviews/AMD/Radeon_RX_Vega_64/10.html
https://www.techpowerup.com/reviews/AMD/Radeon_RX_Vega_64/14.html
https://www.techpowerup.com/reviews/AMD/Radeon_RX_Vega_64/17.html
https://www.techpowerup.com/reviews/AMD/Radeon_RX_Vega_64/18.html

The point is, both cards can overclock and be tweaked to run better than stock. Both cards trade blows, So claiming one benchmark as an indicator is disingenuous at best and outright bias at worst.

I think you missed the point of my post. And if you check my post history you will see I make a very simaler point a few pages back about not relying on just one game.

The point of my above post was quite literally to say why just tweak one when you can tweak the other and get even better results.

As for Vega 64 overclocking... nah

 

[ICDP]

I think you missed the point of my post. And if you check my post history you will see I make a very simaler point a few pages back about not relying on just one game.

The point of my above post was quite literally to say why just tweak one when you can tweak the other and get even better results.

As for Vega 64 overclocking... nah

Lol, I could say the same for 1080

Joking aside the RX 64 can be overclocked quite well. I got mine from sitting at (talking GPU warmup times) ~1420 stock clocks to 1650 on air and the HBM2 overclocked to 1050 This is ~15% OC and I got an almost linear increase in performance of around 12%. More overclocking headroom is available with the AIO versions. It's the same with GTX 1080, most of the overclocking headroom is actually just preventing the cards from throttling. Which on the stock versions is not easy due to poor cooling solutions.

The last genuinely good overclocking GPU in my opinion was Tahiti. 925 core clock on 7970 and could reach 1170-1200, even better with the 7950 which had an 800MHz core clock that could go to 1200.

 

[opethdisciple]

I want AMD to succeed with Vega. But it seems every decision they made was the wrong choice.

Why not just go with GDDR5x and GDDR5 ram like Nvidia. That would have solved a lot of their problems right?

 

[arc@css]

I want AMD to succeed with Vega. But it seems every decision they made was the wrong choice.

Why not just go with GDDR5x and GDDR5 ram like Nvidia. That would have solved a lot of their problems right?

I think the main problem with Vega is how long it has taken for them to release it, and of course, the current pricing situation doesn't help them either.

----------

How's the new card, btw?

 

[Razor Time]

The last genuinely good overclocking GPU in my opinion was Tahiti. 925 core clock on 7970 and could reach 1170-1200, even better with the 7950 which had an 800MHz core clock that could go to 1200.

The 980Ti was a very good overclocker and would go from stock 1,194Mhz to 1,500Mhz or more. My old card hit 1530Mhz.

 

[ICDP]

I want AMD to succeed with Vega. But it seems every decision they made was the wrong choice.

Why not just go with GDDR5x and GDDR5 ram like Nvidia. That would have solved a lot of their problems right?

http://www.gamersnexus.net/guides/3032-vega-56-cost-of-hbm2-and-necessity-to-use-it

No is the simple answer. It may seem the wrong choice but AMD have I believe a large amount of potential with HBM2. HBM2 was meant to be quite a bit faster but due to difficulty's AMD ended up with slower HBM2 nodules. A revised Vega with faster HBM2 would push it a lot closer to 1080Ti speeds.