Displaying items by tag: AMD

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.

AMD can be a confusing company. Over the years they have made more than a few choices that just do not seem to make sense, but we are not going to dive into that right now. Instead we are going to focus on a recent decision that seems both ill-conceived and sure to confuse AMD supporters.

When I first started truly reporting on the tech world there were two power houses making CPUs. You had AMD on one side and Intel on the other. There were a few other companies that were making x86 compatibles, but no one was close to putting out processors in the numbers that AMD and Intel were. Intel was the king with a massive chunk of the market share. After a few deals with DEC and Samsung AMD came up with a CPU and chipset that changed the mind of many enthusiast and started the era of real performance competition between the two.

If you were to look back over the years and pick one seriously mismanaged tech company it would probably be AMD. Even going back into the K6 days we find that AMD just did not know what it wanted to do or what direction it intended to go. The marketing message was confused both internally and to the market as a whole. This confusion hit “Office Space” like proportions in the late 90’s when it was rumored that there were more managers and project leaders than they were actual people doing the work.

When I first started covering the computer world the most common resolution was 640x680 with the hard core gamers getting 800x600. The dream of the day, which some called the golden age of gaming, was 1600x1200 with around 30Fps. Now the dream is “photo realistic” resolutions without the need for heavy anti-aliasing and texture filtering. Even in the mobile world this is becoming a bigger issue with retina displays on the Apple side of the world and 3 and 4k screens on the PC side.

In any technology based company there is a need for strong leadership. The problem comes in finding the right person to take the role of leader. We have seen how much of an impact the right and wrong choices can make in companies like Apple and AMD. As current CEO Rory Read prepares to step down we are taking a look at AMD and their leadership over the years. With AMD they have fought the battle of leaders ever since the tenure of Hector Ruiz. Ruiz is possibly one of the most universally disliked CEOs that AMD has ever had. Most will put AMD’s current situation squarely in his lap.

AMD has been something of an unusual company ever since they first decided to dive into the x86 market with their purchase of NexGen. The would-be CPU maker had an interesting knack of building CPUs that performed well, but were always just a pace behind their rivals. That was the case until AMD pulled off a minor miracle in the form of the Athlon and Athlon64 CPUs. AMD seemed to have stolen the crown from Intel and looked likely to keep it for a long time.

There is a rumor that is claiming AMD will not have a new desktop processor until 2016 that has been making the rounds today. The claim is that AMD is shifting their current focus to mobile CPUs and APUs. Originally it was thought that Carrizo would be the next chip offering that that it would keep up with Intel by offering support for DDR4.

One of the items we have always beat AMD up on is there poor memory performance in their CPUs and APUs. This little issue is what has separated AMD from Intel since the AM2 days. It has always been understood that latency has a massive impact on an internal memory controller. As you latency increases your efficiency decreases. You can offset some of this by enlarging your cache and also optimizing the CPU to use it more efficiently. This is one area that AMD has traditionally had issues with, even going back to the Athlon 64 we saw them reducing cache sizes to remove problems and bump performance.

In the world of silicon size matters, but not in the traditional way. Manufacturers are always looking to make things smaller so they can stuff more into the same space. However there are two types of die shrinks. In the CPU world you see what it commonly referred to as full die shrinks while GPUs tend to do things in half-node steps.