Home Toys Interview
- October 2004 -
[Click Message To Learn More]
I predict in another 5 years the majority of households will have switched to HD. Once you go HD you can never watch SD again. Kind of like being without a microwave, once you have one you can't live without it.
Q1: Is DVI/HDMI/HDCP a connection for the future or on it's way out ... and if so what do you feel will replace it?
DVI/HDMI/HDCP in my opinion is a temporary format. The format was never designed for long distance, does not support networking and there are too many standards making the format a "non standard". I believe IEEE1394B 1.6 GB will in the near future take over with its networking abilities and less restrictive 5C licensing. Its small manufacturer friendly. In the long term, 10 GB Ethernet is likely to be the winner for allowing the ultimate in product convergence, wireless connectivity and GUIs with auto-detect plug and play driver loading. Both IEEE1394 (Firewire) and Ethernet allow drivers to be loaded on power up making both formats very "connective". Turning on a new TV in another room can be auto detected quite easily with little to no programming experience needed by the user. Up to 64 nodes of products makes Firewire a much more attractive format for the "whole house" automation, A/V solution.
Q2: IEEE1394B, 10-Gigabyte Ethernet, and even wireless: How will these upcoming connections change the home theater environment?
I think I addressed this question in the earlier comments on why in my opinion DVI is a temporary format. Wireless has a bit to go to allow HD transmission but it will get there as chip sets continue to be developed.
Q3: What are your thoughts on LCOS, DILA, and DLP? Which would you prefer in your home theater and why?
I like LCOS, always have but the black levels still are not there. Once polarizers are developed especially in blue, blacks will improve but I still think the technology has a bit to go. I like the electronic MTBF (Mean Time Before Failure) curve of LCOS. For a manufacturer, this is easier to stomach than the mechanical one of DLP. As time goes on an LCOS chip is less likely to fail whereas in a DLP it is more likely to since its a mechanical device.
DLP outperforms LCOS or LCD in both dynamics and black levels. The disadvantage of course is the screen door effect and the dithering of pixels especially noticeable in low light levels. This is due to the mirrors modulating at a low rate the eye can perceive. Once the DLPs can achieve a higher modulation rate the frequency , the mirror modulation could be increased to the point the eye would not perceive it. That and the higher resolutions/higher fill factor could and should make the DLP for the near future out-perform LCOS. DILA is LCOS by the way, just the brand name that JVC put on the technology and this sometimes confuses the consumer. Both DLP and LCOS have the potential to produce stunning pictures.
Q4: Which is better and why ... a Built in or Outboard scaler?
Virtually all projectors and most DVD players have built in scalers. Although their performance is "OK", they simply can not match the brute force scaling engines of good external scalers. Now when I say good scalers, Im not talking about inexpensive scalers that basically use the same chip sets as the internal ones but scalers that implement algorithms and innovative techniques that experienced video engineers invent. There are several companies out there, many in the commercial world that continue to pave the way with new and innovative methods to improve PQ. Picture Quality (PQ) improvements are a continuous and persistent journey. You keep chiseling away at problem areas and through many small improvements the PQ continues to improve. It's the persistence and R&D that many companies take the short cut way out. Unless your developing your own techniques, your simply another "me too" manufacturer and although you may be successful commercially, your not contributing to the science of PQ improvement. The passion versus commerce conflict keeps many companies from developing anything new.
Q5: What makes a good video processor?
A good processor needs to first and foremost be flexible. It must be improved as you come up with new ideas to attack problems. A software-based processor has a huge advantage in this area. A great processor should allow its user to download improved algorithms over the Internet as better methods are invented . Areas like Anti Aliasing, MPEG & scaling artifact reduction, Noise reduction, auto lip sync detection and other problem areas are places a good processor should never dead end its user. Of course this type of flexibility doesn't come cheap, so don't expect a product using a $9.00 ASIC to be able to accomplish such flexibility, it can't happen in this type of device. The processor should be FPGA and software intensive, software residing on DROMs, using super fast serial and/or parallel processors. As processor speeds increase, so do the power of the algorithms and PQ improvement techniques.
Q6: Why is the word "investment" not mentioned during the home theater purchase?
This is due to the fact that most products are developed with obsolescence in mind. The product is usually obsolete before it's bought because the manufacturer already has a newer generation product in the lab in prototype. Manufactures need sell through to fund future R&D so no one will ever talk about the fact that the product they just released is already obsolete. It takes a good 12 months to develop a major product, sometimes large companies will take 2 - 3 years to discuss the possibility of making a product. To have a good investment in this industry you need to be able to purchase a product that wont be obsolete the day you purchase it. Usually only small companies can afford to do this. Most large companies depend on manufacturing products in the large quantities. Its the only way to get their board cost down to allow the general public to afford it. A board that may cost a small manufacturer $500.00 in 100 lot quantities may cost a large manufacturer less than $15.00 in 100,000+ lot quantities. Once the 100,000+ are sold out, a new product is already designed to take its place. Obsolescence time is short and Investment is non-existent.
Q7: What makes a home theater product upgradeable ... and why don't most manufactures allow upgrades?
A product has to give its user the ability to keep their initial investment. An upgrade cost should allow the consumer to invest only in the new technology that renews the products performance and features to stay state of the art. A product should continue to stay on time of the technological wave much like a surfer stays on their wave.
Most manufacturers can't make products upgradeable. Imagine the logistics of returning 100,000+ products, upgrading them and returning them to the consumer. It's a nightmare. Why would any large manufacturer do that when they simply can make a product disposable? The product may not be as high performance as that small manufacturer product but who cares? They have their dealers, their distributors, they are pushing cardboard boxes in 100,000+ quantities, share holders are happy, everyone is making money.... Again, the passion VS. commerce conflict. The industry needs the small guys to keep pushing the envelope. In our industry, the "tail wags the dog".
Q8: How long do you think it's going to be before HDTV (or at least digital TV in some format) takes over as the standard for home entertainment systems ... or will it?
I predict in another 5 years the majority of households will have switched to HD. Once you go HD you can never watch SD again. Kind of like being without a microwave, once you have one you can't live without it. I will make the comment that just because the industry goes HD does not mean the passion for better PQ stops. HD has many problems, MPEG artifacts are huge as compression is still implemented and even more so in HD to get the bandwidth out of starved broadcast links. Processors are going to be needed MORE, not less. Many people have asked me what is TAW going to do once HD is so prevalent, scalars and processing won't be needed. They will be needed more as the passion towards a perfect PQ never ends.