Interview with Creative labs Europe -
The Creative Labs Interview
Guru3D.com has interviewed Eoin Leyden, Branch Manager for the Graphics Card Division of Creative Labs Europe. There has been a lot going on surrounding Creative Labs Europe and their graphics card segment, it is now official, Creative Labs Europe is following Terratec and Hercules and pulling out of the consumer graphics card market.
Quite disturbing as Creative Labs has been a big player in this market. Puzzled by what has been happening in the industry I've interviewed him to learn why exactly they are halting their sales of consumer graphics cards. I have known Eoin for quite a while now and he was interviewed several times here at Guru3D.com, I've known him to be a honest and open minded individual, and that's exactly what you'll see in his answers.
Let's get started.
First off, can you tell to the audience who you are and what you do at Creative Labs Europe.
Sure, my name is Eoin Leyden and I work for Creative Labs Europe. Ive spent the past seven years managing Creatives graphics business in Europe, which includes pretty much every aspect of the product life from selecting technology partners and selecting specific product configurations to choosing contract manufacturing sites to actually build the boards.
We recently learned that CL Europe is stepping out of the Graphics Card business (for the time being). Will this effect Europe only, or other CL divisions world-wide also ?
Creative Labs is divided into three regions Europe, Asia, US and while our core products are common to each region (e.g. audio cards, MP3 players etc), each region has autonomy in other areas. In recent years while Europe has maintained a strong retail graphics business our US team decided to drop the retail graphics line and Asia has taken an opportunistic approach and still offers product based on both ATI and NVIDIA chipsets. (As an aside we do ship one graphics product in the US which is the Graphics Blaster Picture Perfect. This is based on a 3D Labs chipset and targeted not at the 3D gamer but more at digital camera owners). Going forward I do not believe Creative Europes decision will have an effect on the strategies of Creative in Asia or the US.
We know that the videocard industry is a harsh one. It's hard to stay competitive, you need to have the right product at the right place at the right time for the right price. I've known you for quite a while now and from what I can tell Creative labs Europe always managed to make dynamic decisions in terms of choosing a graphics chipset manufacturer. Switches from 3dfx, to NVIDIA, to ATi, ect... you do not doubt the market and know what the end-user is willing to buy. Can you explain what it is exactly that is so difficult in this industry ?
One of the fundamental difficulties in this business is selecting the right products to offer the customer. On the surface it should be easy to map out the price performance of each available product and figure out which is the best value for money unfortunately many other factors come into play. It is easiest to explain if I give you an example.
When NVIDIA launched the GeForceFX 5200 I was faced with an unbelievable number of potential implementations. The frame buffer could be 64MB, 128MB or 256MB, the memory interface could be 64-bit or 128-bit, it could have VIVO or DVI connectors. The challenge is to figure out what is important to the customer (and what features they are willing to pay for). So using the above example you get questions like Should we do a 256MB version? On the one hand everyone wants more memory and the additional cost is not huge so the answer should be yes. On the other hand because of the limited power of the 5200 chipset it is unlikely you would ever run apps at 1600x1200 with AA enabled and so you will never actually use the extra memory so the max should be 128MB. Why have customers pay for an extra 128MB they will never use?
And so it goes on all of the available chipsets. Rather than NVIDIA and ATI defining the product offerings they simply sell the chipsets and let the partners choose the actual implementation. This can be good in one way but also bad. The recent issues with the Radeon 9800 Pro 128-bit vs 256-bit that you highlighted on Guru3d is a good example. People are being led to believe they are getting a bargain when in fact they are getting a severely impaired product.
Right now, both NVIDIA and ATi have some extraordinary products on the market, both the x800 and GeForce 6800 series are downright impressive the feedback we receive on these products are amazing, people are willing to spend way more money on their graphics card in the high and mid-end range then ever before. When you look at these new products are you still confident that you are not 'missing the boat' here ?
Not sure where to start on answering this one. Yes they are incredibly powerful products and yes people are paying a lot of money for them but the really important part from a manufacturers point of view is can we make any money selling them and the answer to that is basically no. On these high end cards the designs are extremely complex and most people stick to reference design. This means that ultimately everyone is forced to compete on price and while this is great for the end user it is a major problem for board manufacturers. The graphics business is actually quite expensive to run when you factor in the costs of technical support, returns etc. I think we have clearly seen this over the years by the number of companies who have gone to the wall 3Dfx, Visiontek, Diamond, Elsa etc. etc.
As a follow-up to that question, can you explain a bit more precisely why you are stepping out of the graphics card industry? Does it have to do with the chipset manufacturers and in-between relations ? or simply economics? Is it that hard to earn a buck in the bizz? I'm asking this so precisely as a lot of people simply do not understand what is going on. We've seen Terratec leave the graphics card market, Hercules and now Creative labs. Which are three VERY strong names here in Europe.
This is partly to do with my point above but to explain it more clearly I need to go back a few years. When Creative launched the Voodoo2 we had a large team of hardware and software engineers working on our graphics business. The reason for this was to make our product better than the competition. Along with our own hardware designs we used to do our own control panel and wrote additional software features exclusively for our cards (e.g. Voodoo2 could only address textures of 256x256 we wrote our own code so it could address up to 1024x1024). We continued this with the original Riva TNT where we customized hardware and software to differentiate our products. (I dont know if many people remember but it was around this time we wrote the shadow patch for Unreal which was the first time I remember seeing volumetric shadowing in real time). So at this time there was real differentiation between different graphics providers... a Voodoo2 from Creative was a very different beast than a Voodoo2 from Diamond.
What happened next was that the speed of chip development advanced to the point where we were getting new products every 6 months. Before this you could sell a high-end card for over a year but now the life was more like 4 months. This created a problem for us in that we could not afford to spend time customizing products and doing our own hardware and software development or the product would be obsolete before we got to market. To make matters worse we were no longer competing with just established brand names like Diamond and Elsa but also with a huge number of non-branded vendors from the Far East.
Both NVIDIA and ATI, as mentioned above, have a philosophy of selling the chips to anyone and everyone and letting the manufacturers fight it out among themselves. This approach, coupled with the fact that most people now ship reference design cards (for the reason mentioned above) means that all that is left to compete on is price. In the majority of cases when a market develops to this point nobody makes any money as somebody is always willing to sell product at or below costs in order to gain market share or to clear inventory.
So in summary, we have a combination of factors at work. (1) The number of product variations makes it hard to pick the right product (2) The margins are very thin because of lack of differentiation (3) Any change in the memory market can wipe out any profits you make (4) Even if you get all of this right you could simply be working with the wrong partner because if you miss a technology shift your toast.
You told me that CL Europe is stepping out the graphics industry "for now". What are the chances of a return? At what time-frame are we looking ?
The reason why I say for now is that we will always keep an eye on this business. The demand is still there for 3D and will be for a long time to come. The problem with the category is not that the market is not there but rather it is the business model that ATI and NVIDIA have adopted. By forcing everyone to compete against each other with near identical products they effectively make it impossible for anyone to be profitable. At some point this will have to change and when it does we will look at the market again.
Will the decision have an effect for your professional line of graphics cards, e.g. 3D Labs.
Not at all, 3D Labs are still focused on the workstation and continue to deliver industry leading hardware. The recent Realizm launch is just the latest in a long line of hugely successful workstation solutions.
Warrantee on current products. I assume that people that recently bought a graphics card product from CL Europe will remain having their full warrantee and support for their products ?
Absolutely we will fully support our products and the warranty period will be fully honored.
Eoin, thanks for answering these questions. On behalf of myself I can say that the news saddens me, again a big player in the market who halted selling graphics cards is a huge loss for all of us.
Thanks. It seems like a long time ago since Creative launched the first ever 3D accelerator for the PC (the 3D Blaster VLB) and since then we have worked with just about every chip company out there. Its been great fun developing and marketing these products and Id like to take this opportunity to thank everyone who has supported us over the years.
AMD launched a new high-end graphics card recently. In that spirit we interviewed Dave Baumann. A name that might sound familiar to some of you. Dave Baumann was at the time chief editor for Beyond3D. Back in 2006 he joined ATI (now part of AMD) as technical marketing manager. Dave now is Product manager and both the RV770 and RV790 were his responsibility.
Paccus fluid head force feedback HAWK Joystick interview
An interview with Joseph Melskens from Paccus. A Dutch absed company who is wotrking on the new revelation in Force Feedback based Joysticks, fluid head based force feedback. Paccus developed a totally new kind of joystick. One which is real 3D, is as precise and fast as a game mouse, has extra programmable keys, but most important has a unique (patented) fluid head force feedback, that can pull you out of your seat! This is not a toy, it is made from metal. We also have some photos ogf the prototype.
SiSoftware Sandra 2009 Interview with Adrian Silasi
An interview with SiSoftware Sandra lead programmer Adrian Silasi. Topics involve SANDRA 2009, GPUs and compute API's like CUDA and CAL.
Interview with Neal Robison from ATI
Neal Robison is the person being interviewed today, he's ATIs global Director for Developer Relations and despite his very busy schedule we have managed to get Neal to agree to Q&A session. We talk a little about the HDR & AA, gaming in general, consoles and as we are both in this industry .. the future of gaming.