Over the last couple of days, we have received information that would indicate nVidia is not moving to HBM 2 for their consumer GPUs (outside of some extremely high-end models). Instead, they appear to be focusing on improvements found in GDDR5X and GDDR6. Conversely, AMD appears to be focusing on HBM for many of their high-end and even some mid-range cards. The two very different paths has sparked something of a debate amongst fans of both products (as you can imagine). The questions are, why chose one over the other at this point and is HBM a truly viable option for AMD?
Yesterday we talked about the possibility that AMD will launch a Dual-GPU R9 Fury X card geared for 4k and VR. This is certainly welcome news for most AMD fans and for fans of virtual reality. It was no coincidence that the first time we are seeing this in operation was at a big VR event in LA or that the launch is rumored to coincide with the launch of Oculus and HTC’s Vive headsets. This move would be a very high-end AMD card on the market around April/May of this year.
As most people are aware, AMD dropped the first GPUs to utilize HBM (High Bandwidth Memory). These GPUs use a form of HBM called 2.5D which requires the use of an interproser layer than both the memory and the GPU sit on. This is opposed to the 3D stack in which the memory sits on top of the processor that owns it. The traditional stacking of 3D Memory provides significant performance benefits, but would require a different chip for every memory density you plan on releasing. In the GPU world this can be a big problem and is why both AMD and NVIDIA have opted for the 2.5D method.
In September of 2015 (that would be this year) AMD announced that they were splitting off their GPU business into the Radeon Technologies Group (RTG). This move had a number of reasons (most good) and would serve to distance the graphics group from the CPU business. After talking to a few investors they were very optimistic about this move and would consider investing in RTG where they might not have done so in AMD. The split was a long time coming and is actually how AMD should have handled the ATi buy back in 2006 (Merger with separate business units and not a complete buy out).
Although much of the press surrounding AMD at the moment is focused on their lackluster earnings for Q2, there is some potentially good news from them. AMD’s dive into the use of High Bandwidth Memory is going to continue with their next GPU line up. According to the information available the next generation of GPUs will be code named Artic Islands and should be manufactures on a 16nm FinFET process.
A couple of days ago we published an article on the state of AMD and what their immediate (next 18 months) looked like. In that article we looked at the state that AMD is in right now with what they have on the table…. It seems that a few readers did not like what we had to say. Oddly enough, yesterday a number of articles popped up on the internet that supported much of what we had to say including many of the time lines (14nm by Q3 2016 etc.) Let’s take a look at some of the information out there.
There is an old saying in business that you have to spend money to make money. The meaning is sort of plain, but we will give you a clear example of what it means. In the tech world, if you are not spending money to improve your products or to build the next generation then you are pretty much going to have a very short life. This is not to say you have to dump huge mounds of cash into research and development, but you do have to spend some on the right things. This means that you can tell a lot about a company’s focus and future outlook by simply looking at their R&D spending.
We have talked about some of the challenges that AMD is facing due to poor sales (and bad leadership), but many of these challenges have not been as evident as what is happening in the GPU world. Although we have talked a little bit about AMD’s next generation cards from the top of the heap, we have not really looked at what is happening with the rest of the line.
During NVIDIA’s recent GTC announcements the world was shown the new Titian X with 12GB of GDDR5. This impressive monster of a card has shown that it has a large amount of power to push your games and other graphical information. While the Titian X received adoration and several very positive reviews from the technical press there was another story that was also very important. This was the conversation about NVIDIA’s next GPU, Pascal.
Some GPU news comes today from the AMD camp where we are hearing rumors that AMD’s Radeon 390X might ship with 8GB of High Bandwidth Memory (HBM). Now this sounds great when you take into account the fact that HBM has some serious muscle behind it (128GBps per memory chip). AMD could technically leapfrog nVidia’s new monster the GeForce Titian X with 12GB of memory onboard. The problem is that some of the information that is floating around does not seem to add up.