AMD Radeon HD 7970 Review: 28nm And Graphics Core Next, Together As One
by Ryan Smith on December 22, 2011 12:00 AM EST- Posted in
- GPUs
- AMD
- Radeon
- ATI
- Radeon HD 7000
Video & Movies: The Video Codec Engine, UVD3, & Steady Video 2.0
When Intel introduced the Sandy Bridge architecture one of their big additions was Quick Sync, their name for their hardware H.264 encoder. By combining a specialized fixed function encoder with some GPU-based processing Intel was able to create a small, highly efficient H.264 encoder that had quality that was as good as or better than AMD and NVIDIA’s GPU based encoders that at the same time was 2x to 4x faster and consumed a fraction of the power. Quick Sync made real-time H.264 encoding practical on even low-power devices, and made GPU encoding redundant at the time. AMD of course isn’t one to sit idle, and they have been hard at work at their own implementation of that technology: the Video Codec Engine (VCE).
The introduction of VCE brings up a very interesting point for discussing the organization of AMD. As both a CPU and a GPU company the line between the two divisions and their technologies often blurs, and Fusion has practically made this mandatory. When AMD wants to implement a feature, is it a GPU feature, a CPU feature, or perhaps it’s both? Intel implemented Quick Sync as a CPU company, but does that mean hardware H.264 encoders are a CPU feature? AMD says no. Hardware H.264 encoders are a GPU feature.
As such VCE is being added to the mix from the GPU side, meaning it shows up first here on the Southern Islands series. Fundamentally VCE is very similar to Quick Sync – it’s based on what you can accomplish with the addition of a fixed function encoder – but AMD takes the concept much further to take full advantage of what the compute side of GCN can do. In “Full Mode” VCE behaves exactly like Quick Sync, in which virtually every step of the H.264 encoding process is handled by fixed function hardware. Just like Quick Sync Full Mode is fast and energy efficient. But it doesn’t make significant use of the rest of the GPU.
Hybrid Mode is where AMD takes things a step further, by throwing the compute resources of the GPU back into the mix. In Hybrid Mode only Entropy Encode is handled by fixed function hardware (this being a highly serial process that was ill suited to a GPU) with all the other steps being handled by the flexible hardware of the GPU. The end goal of Hybrid Mode is that as these other steps are well suited to being done on a GPU, Hybrid Mode will be much faster than even the highly optimized fixed function hardware of Full Mode. Full Mode is already faster than real time – Hybrid Mode should be faster yet.
With VCE AMD is also targeting Quick Sync’s weaknesses regardless of the mode used. Quick Sync has limited tuning capabilities which impacts the quality of the resulting encode. AMD is going to offer more tuning capabilities to allow for a wider range of compression quality. We don’t expect that it will be up to the quality standards of X264 and other pure-software encoders that can generate archival quality encodes, but if AMD is right it should be closer to archival quality than Quick Sync was.
The catch right now is that VCE is so new that we can’t test it. The hardware is there and we’re told it works, but the software support for it is lacking as none of AMD’s partners have added support for it yet. On the positive side this means we’ll be able to test it in-depth once the software is ready as opposed to quickly testing it in time for this review, however the downside is that we cannot comment on the speed or quality at this time. Though with the 7970 not launching until next year, there’s time for software support to be worked out before the first Southern Islands card ever goes on sale.
Moving on, while encoding has been significantly overhauled decoding will remain largely the same. AMD doesn’t refer to the Universal Video Decoder on Tahiti as UVD3, but the specifications match UVD3 as we’ve seen on Cayman so we believe it to be the same decoder. The quality may have been slightly improved as AMD is telling us they’ve scored 200 on HQV 2.0 – the last time we scored them they were at 197 – but HQV is a partially subjective benchmark.
Finally, with Southern Islands AMD is introducing Steady Video 2.0, thesuccessor to Steady Video that was introduced with the Llano APU last year. Steady Video 2.0 adds support for interlaced and letter/pillar boxed content, along with a general increase in the effectiveness of the steadying effect. What makes this particularly interesting is that Steady Video implements a new GCN architecture instruction, Quad Sum of Absolute Differences (QSAD), which combines regular SAD operations with alignment operations into a single instruction. As a result AMD can now execute SADs at a much higher rate so long as they can be organized into QSADs, which is one of the principle reasons that AMD was able to improve Steady Video as it’s a SAD-heavy operation. QSAD extends to more than just Steady Video (AMD noted that it’s also good for other image analysis operations), but Steady Video is going to be the premiere use for it.
Facebook
Twitter
RSS
292 Comments
View All Comments
CeriseCogburn - Thursday, March 8, 2012 - link
Interesting, amd finally copied nvidia..." This problem forms the basis of this benchmark, and the NQueen test proves once more that AMD's Radeon HD 7970 tremendously benefits from leaving behind the VLIW architecture in complex workloads. Both the HD 7970 and the GTX 580 are nearly twice as fast as the older Radeons. "
When we show diversity we should also show that amd radeon has been massively crippled for a long time except when "simpleton" was the key to speed. "Superior architecture" actually means "simple and stupid" - hence "fast" at repeating simpleton nothings, but unable to handle "complex tasks".
LOL - the dumb gpu by amd has finally "evolved".
chizow - Thursday, December 22, 2011 - link
....unfortunately its going to be pitted against Kepler for the long haul.There's a lot to like about Southern Islands but I think its going to end up a very similar situation as Evergreen vs. Fermi, where Evergreen released sooner and took the early lead, but Fermi ultimately won the generation. I expect similar with Tahiti holding the lead for the next 3-6 months until Kepler arrives, but Kepler and its refresh parts winning this 28nm generation once they hit the streets.
Overall the performance and changes AMD made with Tahiti look great compared to Northern Islands, but compared to Fermi parts, its just far less impressive. If you already owned an AMD NI or Evergreen part, there'd be a lot of reason to upgrade, but if you own a Fermi generation Nvidia card there's just far less reason to, especially at the asking price.
I do like how AMD opened up the graphics pipeline with Tahiti though, 384-bit bus, 3GB framebuffer, although I wonder if holding steady with ROPs hurts them compared to Kepler. It would've also been interesting to see how the 3GB GTX 580 compared at 2560 since the 1.5GB model tended to struggle even against 2GB NI parts at that resolution.
ravisurdhar - Thursday, December 22, 2011 - link
My thoughts exactly. Can't wait to see what Kepler can do.Also...4+B transistors? mind=blown. I remember when we were ogling over 1B. Moore's law is crazy.... :D
johnpombrio - Wednesday, December 28, 2011 - link
Exactly. If you look at all the changes that AMD did on the card, I would have expected better results: the power consumption decrease with the Radeon 7970 is mainly due to the die shrink to 28nm. NVidia is planning on a die shrink of their existing Fermi architecture before Kepler is released:http://news.softpedia.com/news/Nvidia-Kepler-Is-On...
Another effect of the die shrink is that clock speed usually increases as there is less heat created at the lower voltage needed with a smaller transistor.
The third change that is not revolutionary is the bump of AMD's 7970's memory bus from 384 bits (matching the 580) from the 6970's 256 bits along with 3GB DDR5 memory vs the GTX580's 1.5GB and the 6970's 2GB.
The final non revolutionary change is bumping the number of stream processors by 33% from 1,536 to 2,048.
Again, breaking out my calculator, the 35% bump in the number of stream processors ALONE causes the increase in the change in the benchmark differences between the 7970 and the 6970.
The higher benchmark, however, does not show ANY OTHER large speed bumps that SHOULD HAVE OCCURED due to the increase in the memory bus size, the higher amount of memory, compute performance, texture fill rate, or finally the NEW ARCHITECTURE.
If I add up all the increases in the technology, I would have expected benchmarks in excess of 50-60% over the previous generation. Perhaps I am naive in how much to expect but, hell, a doubling of transistor count should have produced a lot more than a 35% increase. Add the new architecture, smaller die size, and more memory and I am underwhelmed.
CeriseCogburn - Thursday, March 8, 2012 - link
Well, we can wait for their 50%+ driver increase package+ hotfixes - because after reading that it appears they are missing the boat in drivers by a wide margin.Hopefully a few months after Kepler blows them away, and the amd fans finally allow themselves to complain to the proper authorities and not blame it on Nvida, they will finally come through with a "fix" like they did when the amd (lead site review mastas) fans FINALLY complained about crossfire scaling....
KaarlisK - Thursday, December 22, 2011 - link
What is the power consumption with multiple monitors? Previously, you could not downclock GDDR5, so the resulting consumption was horrible.Ryan Smith - Thursday, December 22, 2011 - link
"On that note, for anyone who is curious about idle clockspeeds and power consumption with multiple monitors, it has not changed relative to the 6970. When using a TMDS-type monitor along with any other monitor, AMD has to raise their idle clockspeeds from 350MHz core and 600Mhz memory to 350MHz core and the full 5.5GHz speed for memory, with the power penalty for that being around 30W. Matched timing monitors used exclusively over DisplayPort will continue to be the only way to be able to use multiple monitors without incurring an idle penalty."KaarlisK - Thursday, December 22, 2011 - link
Thank you for actually replying :)I am so sorry for having missed this.
ltcommanderdata - Thursday, December 22, 2011 - link
Great review.Here's hoping that AMD will implement 64-bit FP support across the whole GCN family and not just the top-end model. Seeing AMD's mobile GPUs don't use the highest-end chip, settling for the 2nd highest and lower, there hasn't been 64-bit FP support in AMD mobile GPUs since the Mobility HD4800 series. I'm interested in this because I can then dabble in some 64-bit GPGPU programming on the go. It also has implications for Apple since their iMacs stick to mobile GPUs, so would otherwise be stuck without 64-bit FP support which presumably could be useful for some of their professional apps.
In regards to hardware accelerated Megatexture, is it directly applicable to id Tech 5's OpenGL 3.2 solution? ie. Will id Tech 5 games see an immediate speed-up with no recoding needed? Or does Partially Resident Texture support require a custom AMD specific OpenGL extension? If it's the later, I can't see it going anywhere unless nVidia agrees to make it a multivendor EXT extension.
Ryan Smith - Thursday, December 22, 2011 - link
Games will need to be specifically coded for PRT; it won't benefit any current games. And you are correct in that it will require and AMD OpenGL extension to use (it won't be accessible from D3D at this time).