Guru3D.com
  • HOME
  • NEWS
    • Channels
    • Archive
  • DOWNLOADS
    • New Downloads
    • Categories
    • Archive
  • GAME REVIEWS
  • ARTICLES
    • Rig of the Month
    • Join ROTM
    • PC Buyers Guide
    • Guru3D VGA Charts
    • Editorials
    • Dated content
  • HARDWARE REVIEWS
    • Videocards
    • Processors
    • Audio
    • Motherboards
    • Memory and Flash
    • SSD Storage
    • Chassis
    • Media Players
    • Power Supply
    • Laptop and Mobile
    • Smartphone
    • Networking
    • Keyboard Mouse
    • Cooling
    • Search articles
    • Knowledgebase
    • More Categories
  • FORUMS
  • NEWSLETTER
  • CONTACT

New Reviews
The Callisto Protocol: PC graphics benchmarks
G.Skill TridentZ 5 RGB 6800 MHz CL34 DDR5 review
Be Quiet! Dark Power 13 - 1000W PSU Review
Palit GeForce RTX 4080 GamingPRO OC review
Core i9 13900K DDR5 7200 MHz (+memory scaling) review
Seasonic Prime Titanium TX-1300 (1300W PSU) review
F1 2022: PC graphics performance benchmark review
MSI Clutch GM31 Lightweight​ (+Wireless) mice review
AMD Ryzen 9 7900 processor review
AMD Ryzen 7 7700 processor review

New Downloads
Corsair Utility Engine Download (iCUE) Download v4.33.138
CPU-Z download v2.04
Intel ARC graphics Driver Download Version: 31.0.101.4090
AMD Radeon Software Adrenalin 23.1.2 (RX 7900) download
GeForce 528.24 WHQL driver download
Display Driver Uninstaller Download version 18.0.6.0
Download Intel network driver package 27.8
ReShade download v5.6.0
Media Player Classic - Home Cinema v2.0.0 Download
HWiNFO Download v7.36


New Forum Topics
MSI Interrupt Steering Causing Input Lag Windows 10 Info Zone - gEngines, Ray Tracing, DLSS, DLAA, TSR, FSR, XeSS, DLDSR etc. Amernime Zone AMD Software: Adrenalin / Pro Driver - Release Discovery 22.12.2 WHQL Rumor: Further GeForce RTX 4090 Ti specs emerge RDNA3 RX7000 Seriess! Owners Thread, Tests, Benchmarks, Screenshots, Overclocks, & Tweaks! FSR Thread MSI Afterburner showing wrong CPU data? Installing old 20.4.2 Adrenaline on 6650 XT? Forspoken implements Microsoft's DirectStorage API, faster load times; lowers FPS; raises FPS? Cyberpunk 2077 NVIDIA DLSS 3 Update Is Out Now




Guru3D.com » News » AMD Radeon RX Vega 10 chips differ - physically and quite significantly (Updated)

AMD Radeon RX Vega 10 chips differ - physically and quite significantly (Updated)

by Hilbert Hagedoorn on: 08/17/2017 07:29 AM | source: | 31 comment(s)
AMD Radeon RX Vega 10 chips differ - physically and quite significantly (Updated)

Alright, intriguing story and observation. Try to follow me. It seems that Radeon RX Vega 10 GPUs used on Radeon RX Vega 64 differ a little, I’ll step that up a notch, the seems to even be differing a little bit in height. And that might be relevant for the AIB cards.

Earlier on in our forums a user already asked if we had a different sample, he noticed that the GPU looks different compared to some other websites.  After inspecting our review kit dummy sample and the one used on our graphics card sample I noticed it definitely differ. After further inspection I also noticed that the GPU that is used on our Vega 64 sample, differs from other reviewers. Now here's the thing, my Dutch colleagues over at Tweakers.net noticed something similar.  It seems that AMD fabbing Vega at two different locations. And there is a significant enough physical difference:

 

 
Have a look at the above photo, that’s one of the photos I made of the dummy (but real Vega10 GPU) GPU that AMD submitted as a bit of a fun teaser. See that black coating in-between the chips? Now look at the photo below.

 

 
Compare the two, this photo I made from our Vega 64 graphics card sample, it’s just not the same is it? There should be a dark substrate in-between the HBM2 memory as you can clearly see on the upper photo, the 'dark stuff' is clearly missing on the second (pink area).
 

 
Now check out the upper photo, this one is courtesy from our colleagues at TPU, you can see that their Vega 64 sample is using the first model, totally different from my sample. And now finally, take a peek at the next one:
 

 
Now this photo is the very interesting, it is the dummy GPU from Dutch websites Tweakers and Hardware.info combined (photo courtesy goes out to them). Both websites share the same office and have the same boss, but both received a sample kit. By comparing the two GPUs directly towards each other you can see there is a bit of a height difference? It looks like the HBM2 is sitting a slight notch lower compared toward the GPU die right?

I have no idea what the tolerances are or that the angle of the photos trick the eyes a bit, but this is going to be interesting for the board partners as that difference might have an effect on cooling and in the end tweaking performance. The difference is so incredibly small that a bit of thermal compound likely would make this observation irrelevant. Also it simply does not make sense that HBM2 on part A is sitting higher or lower compared to part B, as in the end it's the same materials used.

But the fact remains, the GPU packages do differ.

Update: We've received word back from AMD, I'll just leave that here:

We have multiple manufacturing partners for Vega10 package assembly. To meet the high-volume production that we target with Vega10, we need access to most of the world capacity for 2.5D interposer technology. Each partner has unique processes, including one which has an epoxy fill to level the entire SOC surface. This specificity is inconsequential to the system solution and all Vega10 packages are form, fit and function equivalent. As there is no functional difference, any Vega10 package flavor can be binned into a Vega10 XL, XT or XTX. 



AMD Radeon RX Vega 10 chips differ - physically and quite significantly (Updated) AMD Radeon RX Vega 10 chips differ - physically and quite significantly (Updated) AMD Radeon RX Vega 10 chips differ - physically and quite significantly (Updated)




« Lian Li PC-Q39 Tempered Glass Mini-ITX Tower · AMD Radeon RX Vega 10 chips differ - physically and quite significantly (Updated) · Broadcom announces 802.11ax WIFI chips - Going for 1.8 Gbit/s »

Related Stories

AMD Radeon Vega 17.8.1 Beta 6 driver download - 08/14/2017 04:23 PM
You can now download Radeon Vega 17.8.1 Beta 6 driver download. Attention, this beta revision 6 driver is intended for RX Vega graphics card only. Download here....

Review: AMD Radeon RX Vega 56 8GB - 08/14/2017 03:00 PM
We review that Radeon RX Vega 56 with 8GB graphics memory. At 399 bucks this is the what could be affordable AMD graphics card in that new Vega range. It quite honestly might actually be the better pr...

Review: AMD Radeon RX Vega 64 8GB - 08/14/2017 02:59 PM
The 4096 Shader processor enabled Radeon RX Vega 64 with 8GB graphics memory is being reviewed in this article. The new AMD graphics card has arrived, is released and we'll take you through the archi...

EK Launches Full Cover water blocks for AMD Radeon RX Vega - 08/03/2017 09:20 AM
EK is presenting Full Cover water blocks for the long awaited AMD Radeon Vega architecture based high-end graphics cards. Customers will be able to transform their GPU into a stunningly beautiful sin...

AMD Radeon RX Vega (Consumer version) Spotted in 3DMark FireStrike - 07/25/2017 03:20 PM
Performance figures of a Radeon RX Vega have surfaced in 3Dmark, this is not the Frontier edition, but looks to be the one of the consumer models to be released and announced (hopefully next week)....


7 pages 1 2 3 4 > »


Agent-A01
Senior Member



Posts: 11616
Joined: 2010-12-27

#5463671 Posted on: 08/16/2017 03:47 PM
Looks like they skipped a process in assembling the dies to conserve time and money?

NewTRUMP Order
Senior Member



Posts: 665
Joined: 2017-02-04

#5463673 Posted on: 08/16/2017 03:49 PM
Lets see if my prejudices' towards Chinese manufacturing is wrong. The top sample looks neat and complete and gives off a polished manufacturing look. The second looks like my teenage cousin made it in his basement. By looks and prejudice, the top one I would expect to be from Taiwan and the bottom shabby looking one from China. Where were they manufactured? Both from Taiwan? Well I was honest up front and will admit I was wrong in my assumption. :eyes: In reality the cpu doesn't know that it is ugly and works just as well as the "pretty" one. I guess down the road if there is a rash of doa cards be interesting to do a post mortem on them to see if the fit and finish was to blame.

Hilbert Hagedoorn
Don Vito Corleone



Posts: 45526
Joined: 2000-02-22

#5463680 Posted on: 08/16/2017 03:58 PM
Lets see if my prejudices' towards Chinese manufacturing is wrong. The top sample looks neat and complete and gives off a polished manufacturing look. The second looks like my teenage cousin made it in his basement. By looks and prejudice, the top one I would expect to be from Taiwan and the bottom shabby looking one from China. Where were they manufactured?


No, after further checking both models do originate from Taiwan.

gianluca
Member



Posts: 77
Joined: 2016-07-20

#5463686 Posted on: 08/16/2017 04:13 PM
To me it seems that the two gpus don't have differences in height, just the missing black finiture.
Would be interesting to see if this missing black substrate causes some differences in temperatures or in performance, somehow. I mean, if it's there is for a reason.
Or maybe it's just some insulating material placed there to avoid thermal paste to go in between the elements.

Evildead666
Senior Member



Posts: 1309
Joined: 2003-09-14

#5463689 Posted on: 08/16/2017 04:19 PM
I read something about this yesterday at home.
An Article about the packing that had to be done to protect the Interposer and chips.
As much as I read, there should be no chips without the packing to protect the HBM2 and the GPU Die.
I'll try and find it (i'm at work) if not, i'll look in my history tonight.

7 pages 1 2 3 4 > »


Post New Comment
Click here to post a comment for this news story on the message forum.


Guru3D.com © 2023