Thursday, May 4th 2017

AMD Works on At Least Three Radeon RX Vega SKUs, Slowest Faster than GTX 1070?

by
btarunr
 Discuss (81 Comments)
AMD could be working on at least three SKUs based on its upcoming "Vega 10" silicon to make up its Radeon RX Vega series. Leaked 3DMark validations point to a device ID that's third in a series of possible device IDs of graphics cards based on the "Vega 10" silicon, the 687F:C1, 687F:C2, and 687F:C3. All three SKUs feature 8 GB of HBM2 memory, and according to leaked 3DMark TimeSpy scores, the "slowest" SKU is faster than NVIDIA GeForce GTX 1070. The fastest SKU is in the same league as the GTX 1080 Ti.

The three SKUs could differ with core-configuration and clock speeds. AMD carved four SKUs out of its "Fiji" silicon, the liquid-cooled R9 Fury X, the air-cooled R9 Fury (with 12.5% fewer shaders), the SFF-friendly R9 Nano (full core-config, but aggressive power-management), and the halo dual-GPU Radeon Pro Duo (1st gen). AMD could take a similar approach with "Vega 10." AMD is expected to launch its Radeon RX Vega series within Q2-2017.
Source: Guru3D Forums

Related News

Add your own comment

81 Comments on AMD Works on At Least Three Radeon RX Vega SKUs, Slowest Faster than GTX 1070?

#26
XiGMAKiD
Caring1Who can forget Chevy Vega
Or this Vega
Posted on Reply
#27
Gasaraki
Solidstate89I thought the C1, C2, and C3 indicated the stepping, with C3 being the closest to (or actual) production?

Am I wrong on that?
Ding, Ding, Ding. We have a winner!
Posted on Reply
#28
Doctor Who
NordicI would be disappointed if the 1080ti competitor was a dual gpu card.
I hope not, and seeing how VEGA's architecture was completely overhauled, its very unnecessary for a Dual GPU card. VEGA has more than enough horse power to match & beat the 1080Ti. IMO
Posted on Reply
#29
Steevo
GasarakiDing, Ding, Ding. We have a winner!
Die respins are usually A1, B1, C1

These are probably cut down or salvage parts or dies that didn't make the cut on clock speed, so why not make a new SKY.

AMD has been on point with engineering with a specific target speed and power consumption in mind and making that happen, there is little to believe differently about Vega.
Posted on Reply
#30
Doctor Who
intelzenas expected - a GTX 1070 performance level - and Vega price will determine if it will be good (for example a gtx 1070 level Vega for gtx 1060 price is a winner all day long, but a a little higher than gtx 1070 performance level Vega for price of GTX 1080 - is a looser all day long). it was clear that Vega was nothing that could beat gtx 1080 even year ago (when AMD showed first cherry picked benches), but one thing that bothers me now is this RX 580 rebrand thing - why AMD did released it? when they will release a roughly 40% more powerful and efficient Vega for same price point just few weeks later - there is only two options: A) Vega will be priced uncompetitive (gtx 1070 level of performance for full gtx 1070 price - just something for hyped fanbois to be happy about) or B) Vega release is more than 3 months away (no matter paperlaunches)... and I do not like any of these
You do realize the 1070, 1080 & 1080Ti are overpriced right? GPUs should never cost so much. It's ridiculously over priced. Hopefully AMD doesn't make the same mistake over pricing VEGA, regardless of how fast it is.
Posted on Reply
#31
mroofie
Super XPI hope not, and seeing how VEGA's architecture was completely overhauled, its very unnecessary for a Dual GPU card. VEGA has more than enough horse power to match & beat the 1080Ti. IMO
And you this how exactly?
Posted on Reply
#32
Nabarun
Super XPYou do realize the 1070, 1080 & 1080Ti are overpriced right? GPUs should never cost so much. It's ridiculously over priced. Hopefully AMD doesn't make the same mistake over pricing VEGA, regardless of how fast it is.
All of nvidia's cards are overpriced. So what? They CAN afford to get away with it. Because AMD CANNOT afford to make theirs significantly cheaper. And unless these "at par with nvidia's" are priced aggressively lower than the competition, I don't see the point of wasting money on them which are getting shittier in terms of efficiency compared to nvidia. Just my thoughts - no fanboyism here.
Posted on Reply
#33
Unregistered
MrGeniusOh...is it suddenly ok to discuss this topic now? Really? Says who? Double standard much?
Exactly this. I started a thread, got flamed by fanboys and that was closed. Double standard at the best.

On the matter, guess the safest way to rationally discuss VEGA without drawing flame from AMDFanboy is to say "Let's just wait for the rest of Q2 2017 until it comes"
NabarunAll of nvidia's cards are overpriced. So what? They CAN afford to get away with it. Because AMD CANNOT afford to make theirs significantly cheaper. And unless these "at par with nvidia's" are priced aggressively lower than the competition, I don't see the point of wasting money on them which are getting shittier in terms of efficiency compared to nvidia. Just my thoughts - no fanboyism here.
I agree. Being late to the top performance tier definitely has its disadvantage. Let's hope competition can drive down the price on both camps.
erockerWhen was it not okay?
When you closed my thread? Ring a bell?
#34
64K
xkm1948Exactly this. I started a thread, got flamed by fanboys and that was closed. Double standard at the best.

On the matter, guess the safest way to rationally discuss VEGA without drawing flame from AMDFanboy is to say "Let's just wait for the rest of Q2 2017 until it comes"




I agree. Being late to the top performance tier definitely has its disadvantage. Let's hope competition can drive down the price on both camps.



When you closed my thread? Ring a bell?
You might not mean to but lately you have been coming across as somewhat bitter about AMD because of your experience with the R9 Fury X. People are excited about what Vega might bring and too much negativity kind of spoils the fun. :)
Posted on Reply
#35
cdawall
where the hell are my stars
mroofieAnd you this how exactly?
Shader and core count would point to it as long as clock speed is high enough. Remember a pair of 480's are exactly the same core/shader count. Those equal a 1080 (both at stock clocks) when xfire actually works.

If one were to lets say assume that all of the rumors and heresy from AMD announcing that this was an improved design, well it doesn't take a rocket scientist to go 10% performance bump and 1500mhz would easily equal a 1080ti.
Posted on Reply
#36
Unregistered
More information from CHH again. Takes this with whatever exotic salt you want

chiphell.com/forum.php?mod=viewthread&tid=1731990&page=1&authorid=42930


Google Translate of the leak
wjm47196Pcb length with 480 public version of the same long 64cu to determine

8g and 16g two versions

Make sure there is water cooled version

Originally released in April is currently pushed to June but last month in the pcp trial production at present all the packaging materials to promote the web page manual are not ready to think that you will not delay it

Also computex certainly have the exact message should be officially released but not up to release I personally think there should be more than 90% chance

Say that again 1.2g vega run points is true
#38
FordGT90Concept
"I go fast!1!11!1!"
We've already discussed that first picture on another thread. It's not in Furmark's database therefore, it's likely a hoax. The second picture is in Furmark's database.

Here's the closest to the fake (1080 Ti): www.3dmark.com/spy/1595952

Here's the legit one (with really low clocks): www.3dmark.com/spy/1544741


Edit: The SiSoftware GPU test puts it between R9 290 and R9 290X. That's consistent with the low score TimeSpy benchmark above.

The very big question is what is the reference clockspeed going to be? 1200 MHz? 1525 MHz? 1800 MHz? If only 1200 MHz then everyone is going to be disappointed. If 1800 MHz, we likely have a Titan Xp slayer on our hands.

The same question was asked of Polaris before it debuted. We were anticipating 1800 MHz so it could compete with GTX 1080. What we got was 1266 MHz which didn't compete with GTX 1070, nevermind GTX 1080. Because of this, I'm expecting reference clock will be ~1200 MHz. I hope AMD surprises.
Posted on Reply
#39
Fx
btarunrYou forgot Lou Vega Mambo #5.
Haha. Good beat.
Posted on Reply
#40
efikkan
It surprises me that the author didn't check the sources better.
We already know 687F:C1 is the early ES AMD has demostrated a few times already:
Posted on Reply
#41
nadrojcote
Wasn't it already confirmed that this first picture is not reputable? Surprised to see this article on here. So much speculation going on. Can't wait for real news.
Posted on Reply
#42
kruk
Here is how the hype circle works:

1.) a Reddit user A mistranslates a forum post on a Chinese website and creates a thread
2.) a Reddit reader B posts this thread on a English forum, but adds to confusion by applying wrong text/data corrections
3.) a News writer C which monitors forums for leaks, writes a news article without checking sources. He adds a tiny pinch of hype to get more clicks
4.) thread from Reddit user A is by now on page 2 of subreddit because of downvotes/lack of interest and thus forgotten
5.) a Reddit user D creates a thread referencing news from News writer C
6.) continue from point 2.)

:D

*A, B, C and D represent different persons.
Posted on Reply
#43
Casecutter
fore1gnDon't hope that Vega will start at GTX 1060 price. It will be around $349 launch price, with aftermarket cards going as high as $400+. Middle Vega will probably be $499 and high-end Vega $649. Low- and mid-Vega will be beating the competition, but high-end Vega won't. /speculation
Yes while AMD/RTG is not in cadence step with the competition they aren't running a charity... It has only been lately Nvidia (and in the same vane Intel) got real about 1070 & 1080 pricing; remember the FE model where $450/$700; with aftermarket minimums $380/$600 and not many where near that.

Edit... so yes in agreement.
Posted on Reply
#44
SkOrPn
FrickThat has been going on since Polaris.
That has been going on since technology
Posted on Reply
#45
wiak
btarunrYou forgot Lou Vega Mambo #5.
dammit now i got it back in my head after 18 years

Posted on Reply
#46
Lionheart
btarunrYou forgot Lou Vega Mambo #5.


Brilliant. :clap:
Posted on Reply
#47
Doctor Who
1586 MHz VEGA was Leaked by a so called poster that claims he seen the real VEGA in action behind close doors. On wccftech.com/

Am I surprise by such posts? Lol
Posted on Reply
#48
cdawall
where the hell are my stars
Super XP1586 MHz VEGA was Leaked by a so called poster that claims he seen the real VEGA in action behind close doors. On wccftech.com/

Am I surprise by such posts? Lol
That would show a massive improvement in the process. I mean the BEST RX480/580's can hit 15xx with an overclock on highend air and water. To release a card at stock with those numbers would be top 5% of 480/580 dies. This die is even larger...I cannot imagine the binning that would be required with current tech
Posted on Reply
#49
FordGT90Concept
"I go fast!1!11!1!"
Vega apparently has a longer pipeline so higher clock speeds are attainable.
Posted on Reply
#50
uuuaaaaaa
cdawallThat would show a massive improvement in the process. I mean the BEST RX480/580's can hit 15xx with an overclock on highend air and water. To release a card at stock with those numbers would be top 5% of 480/580 dies. This die is even larger...I cannot imagine the binning that would be required with current tech
Isn't the shader density lower than polaris? I mean if that's the case in theory the thermals should be more manageable...
Posted on Reply
Add your own comment
Nov 11th, 2026 06:31 CST change timezone

Latest GPU Drivers

New Forum Posts

Popular Reviews

TPU on YouTube

Controversial News Posts