IEEE 1394 vs. DVI
Dear Gary: I read with great interest Bob Perry’s article comparing IEEE 1394 and DVI. However, I must say that I think you do your readers a disservice by printing an article written by a major stakeholder in one of the two technologies compared. A better choice would have been a debate between proponents of both technologies, or an analysis by an impartial author. As a software engineer who moved from the computer industry to the consumer electronics industry several years ago, I believe my experience in both realms may help shed light on issues glossed over in Mr. Perry’s article. In the IEEE 1394 + HAVi world, the quality of the user interface that can be presented by “any” product is limited both by the features of HAVi and also by the quality of the HAVi implementation in “each” display device. If I wish to create a consumer product with a rich user interface (like a PVR—Personal Video Recorder) or stringent performance requirements (like a video game), I may not be able to take advantage of technologies like graphics acceleration, video compositing/scaling, picture-in-picture, etc. that I would otherwise use. Furthermore, because the display device is responsible for the actual rendering and performance of the user interface, I, as a designer, must test my product with every conceivable display device to insure proper appearance and proper function, and in large measure, I must design to the lowest common denominator. This is a nightmare scenario that threatens to introduce PC-like compatibility problems to the home entertainment space. Finally, consider this: the display device is likely to be one of the most expensive components of the system, and one of the longest-lived. We’ve all seen the rate at which technology is changing and evolving in the computer and CE industries. Do you, as a consumer, relish a future where the features of your home multimedia system are limited by the capabilities of the HAVi display engine in your six-year-old display? I agree that IEEE 1394 is an excellent recording and intercommunication interface, but I don’t believe it is an appropriate display interface. What we need to produce the kind of high-quality, compelling products that will help drive the DTV transition is IEEE 1394 “plus” a high-quality display interface like DVI.
Michael J. Kobb, Belmont, California
Mitsubishi’s Bob Perry Comments
Mr. Kobb is correct that different displays and source devices may have different capabilities, however that is true today. There is no assurance that a display will properly render an image correctly regardless of source. A user interface on a STB or source device planned for a 720p x 1280 display will appear markedly different on a 1080i x 1920 display, as well as on a 480p x 853 display. HAVi and IEEE 1394 are different, and require different planning and programming. Different is not better or worse, just different. Real gaming machines should have little need for a compressed copy protected interface, and therefore can use component video or RGB, for which connectors exist on all of today’s high-resolution TVs. In fact, to compress the signal just to pass it on a IEEE 1394 connection is wasted cost. In addition, PVR graphical user interfaces are not normally developed at HD resolution, but instead are created for 480p display. A 480p resolution is well within the capability of most HAVi products (EDTV (Extended -Definition TV) is probably minimum display resolution) using both DDI (device driver interface) as well as Level II User Interface. Most designers that create Level II User Interfaces also include DDI software, so that their device is capable of transporting the proper interface to the display product. Having to test products with display devices is a scenario that exists today, and will continue forever. DVI, which comes with baggage such as denying all recording also requires the source box to “adjust” its resolution down to match the display. I agree with Mr. Kobb’s comment that a wonderful world would be both a IEEE 1394, and a variety of uncompressed interfaces such as DVI, Component Video, and RGB for example, for maximum consumer flexibility. Unfortunately, because of selectable output control (the ability sought by Hollywood to “turn off" recordable interfaces), moving to DVI means that in many cases only a DVI signal may be available. This eliminates home recording, and also strands those consumers who placed their faith in manufacturers and the government by transitioning to HD displays early. Mitsubishi cannot accept any public policy which strands our consumers.
You can E-mail Widescreen Review @ mailto:editorgary@widescreenreview.com