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
Fractal Design Pop Air RGB Black TG review
Palit GeForce GTX 1630 4GB Dual review
FSP Dagger Pro (850W PSU) review
Razer Leviathan V2 gaming soundbar review
Guru3D NVMe Thermal Test - the heatsink vs. performance
EnGenius ECW220S 2x2 Cloud Access Point review
Alphacool Eisbaer Aurora HPE 360 LCS cooler review
Noctua NH-D12L CPU Cooler Review
Silicon Power XPOWER XS70 1TB NVMe SSD Review
Hyte Y60 chassis review

New Downloads
Intel ARC graphics Driver Download Version: 30.0.101.1743
AMD Radeon Software Adrenalin 22.6.1 WHQL driver download
GeForce 516.59 WHQL driver download
Media Player Classic - Home Cinema v1.9.22 Download
AMD Chipset Drivers Download v4.06.10.651
CrystalDiskInfo 8.17 Download
AMD Radeon Software Adrenalin 22.6.1 Windows 7 driver download
ReShade download v5.2.2
HWiNFO Download v7.26
7-Zip v22.00 Download


New Forum Topics
AMD is planning to release Ryzen 7000 CPUs in September Nvidia shows signs ... AMD Radeon Software Adrenalin 22.6.1 - Driver download and discussion AMD falls victim of the RansomHouse Extortion Group - 56GB of data stolen AMD Radeon Software - UWP NVIDIA GeForce 516.59 WHQL driver download & Discussion The 13th Generation Raptor Lake ES CPU from Intel is Benchmarked Extreme 4-Way Sli Tuning [3rd-Party Driver] Amernime Zone Radeon Insight 22.5.1 WHQL Driver Pack (Released) According to Asus and Gigabyte, motherboard sales will fall by 25% this year.




Guru3D.com » News » Will AMD Launch a new Single-GPU Card This Summer against GTX 780 Ti ?

Will AMD Launch a new Single-GPU Card This Summer against GTX 780 Ti ?

by Hilbert Hagedoorn on: 05/12/2014 09:09 AM | source: | 20 comment(s)
Will AMD Launch a new Single-GPU Card This Summer against GTX 780 Ti ?

Over the weekend a lot of chatter started on the web and in our forums as at several sources report a new AMD SKU being in development. Basically AMD would be set to release a new graphics card that is lined up against the GTX 780 Ti to compete even better. The new product would use what is called HBM (high-bandwidth memory).

 This is technology that uses multiple stacked DRAM dies with dedicated memory paths into a single package, cutting down on power-draw, thermals, and PCB materials. There three HBM memories availabke, 2Hi, 4Hi and 8Hi.

Now what we 'could' expect is say a product from the Volcanic Islands line being utilized with the first generation of HBM. The first variant to launch soon is 4Hi, which has a bandwidth of 128 GBps and four layers of DRAMs. Each module has 1GB of memory, so if you had 4 of them, then your bandwidth would be 512 GB/s.

The intersting thing is that a leaked document kind of shows the roadmap of future AMD GPUs. Thatb however is a complete shot in the dark. Volcanic Islands would be replaced with Pirate Islands somewhere in the second half of next year. Pirate Islands will be the first 20nm series. This basically means we are more than a year from now till 20nm node arrives from AMD. The 20nm process will not be our guest for long, AMD will quickly move forward to 14nm process with Pirate Islands Refresh.

  • 28nm TSMC 1H 2014 : Hawaii, VI 1.0
  • 28nm GlobalFoundries 2H 2014 : Iceland and Tonga, VI 2.0
  • 28nm GlobalFoundries 1H 2015 : Maui, VI 2.0
  • 20nm GlobalFoundries 2H 2015 : Fiji and Treasure, PI 1.0
  • 20nm GlobalFoundries 1H 2016 : Bermuda, PI 1.0
  • 14nm GlobalFoundries 2H 2016 : Mid-GPU and Low-GPU, PI 2.0
  • 14nm GlobalFoundries 1H 2017 : High-GPU, PI 2.0
To date these are all rumors, but the noise on this is so hard that there has to be some truth to this. And hey, look at the screenshots. As for the results of a product with HBM memory .. we don't know. We do know that the R9 series already have a truckload of memory bandwidth and that the current same memory bandwidth is not a bottleneck at all.



Will AMD Launch a new Single-GPU Card This Summer against GTX 780 Ti ? Will AMD Launch a new Single-GPU Card This Summer against GTX 780 Ti ? Will AMD Launch a new Single-GPU Card This Summer against GTX 780 Ti ? Will AMD Launch a new Single-GPU Card This Summer against GTX 780 Ti ?




« Craven Marsh - CRYENGINE Real-Time Short Film · Will AMD Launch a new Single-GPU Card This Summer against GTX 780 Ti ? · ECS Releases Intel 9-Series L337 Gaming Motherboard »

4 pages 1 2 3 4


kd7
Senior Member



Posts: 151
Joined: 2014-03-22

#4815938 Posted on: 05/11/2014 03:22 PM
beat you to it :p

http://forums.guru3d.com/showthread.php?t=389066

LongWayRunner
Unregistered



#4815942 Posted on: 05/11/2014 03:30 PM
What does HMB exacly achieve?

And how much RAM does the card in the chard have?

kd7
Senior Member



Posts: 151
Joined: 2014-03-22

#4815943 Posted on: 05/11/2014 03:36 PM
What does HMB exacly achieve?

And how much RAM does the card in the chard have?

3D stacking manufacturing process, GPU and memory chips are closely stacked and thus better (shorter) connectivity between them => high bandwidth with lower power.

with first generation of HBM you'll get 2GB at 256GB/s or 4GB at 512GB/s / core. Later versions can scale up to 8GB/core at 2TB/s (2016).

Edit: Actually it is up to 64GB at 2TB/s with 2nd generation HBM.

4xHBM Gen 1 = 4 GB
3xHBM Gen 1 = 3 GB
2xHBM Gen 1 = 2 GB
1xHBM Gen 1 = 1 GB

4xHBM Gen 2 = 16 GB/32 GB/64 GB
3xHBM Gen 2 = 12 GB/24 GB/48 GB
2xHBM Gen 2 = 8 GB/16 GB/32 GB
1xHBM Gen 2 = 4 GB/8 GB/16 GB

http://www.overclock.net/t/1488641/next-r9-r7-graphics-to-use-hbm

LongWayRunner
Unregistered



#4815946 Posted on: 05/11/2014 03:47 PM
3D stacking manufacturing process, GPU and memory chips are closely stacked and thus better (shorter) connectivity between them => high bandwidth with lower power.

with first generation of HBM you'll get 2GB at 256GB/s or 4GB at 512GB/s / core. Later versions can scale up to 8GB/core at 2TB/s (2016).

Thanks a lot!

Also, can they do 8GB at 512 bit with the 1st gen HBM or only 4GB is allowed?

kd7
Senior Member



Posts: 151
Joined: 2014-03-22

#4815949 Posted on: 05/11/2014 03:53 PM
From these leaked slides it doesn't look like, they show a limit of 4 stacks/128GB/s (4x128=512) with 1GB/stack maximum for the first generation.

In fact the HBM bus width is 1024bit even for the first generation, so make that 1TB/s

4 pages 1 2 3 4


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


Guru3D.com © 2022