> But again, if the performance of the tuner is okay, then > the focus of increased CMOS density does not need to > be to make the tuner part acceptable. If you are not continually improving your tuner performance, I will eventually get your business... :-) > So, if you increase density to add in more features, e.g. > to include the MPEG decoding or other features in the > same chip, then overall cost WILL go down. Greater > circuit integration -> lower production cost -> lower > price. As long as you are below the knee in the yield curve, otherwise a 2-chip solution can be cheaper. Or if one of the chips has a lower yield, you don't want to drag down the yield of the bigger chip. BTW, MPEG decoding is a small percentage of the die area of many HDTV decoder chips. Anyway, in general you are correct -- I just can't resist responding to blanket statements... :-) > To say that higher CMOS density results in higher > temperatures and more power is not true in and of itself. > It only happens that way with PCs because guess what, the > clock rate climbs every time the CMOS get denser. But > that's just because the goal with PCs is still faster > and more powerful CPUs. Even the CPUs in chips for HDTVs have to keep going faster due to more complex middleware and I/O requirements. > > You also have to look at the cost/performance/power > > curve for going to 0.13 or 0.09. Power may go down, but > > cost could go up. >=20 > If that were true, today's PCs would cost a heck of a lot > more than my ancient $4000 IBM PS/2 with 80286 chip. OK, momentary cost increase due to using new technology. > Clearly, there's more to the cost equation than merely > the startup cost of going to a new die. Mask charges, wafer costs and yields... > The simple fact is, within a short amount of time, this > 8-VSB vs COFDM debate will be history. The only thing > that amazes me is why this is so hard to acdept. I've > been saying this ever since the RCA DTC-100, 1st gen > receiver, went for upwards of $550. It was the first > ATSC receiver to break the $1000 barrier. So-so > performance and a stiff price. >=20 > Since those days, performance has improved bit by bit, > and price has dropped. With the 4th gen Linx, we finally > went around that second bend in the S curve, where > performance was becoming acceptable. Now we see prices > under $300, and predictions for STBs at $67. What's so > hard to understand? Nothing - we are integrating as fast as we can... :-) ---------------------------------------------------------------------- You can UNSUBSCRIBE from the OpenDTV list in two ways: - Using the UNSUBSCRIBE command in your user configuration settings at FreeLists.org - By sending a message to: opendtv-request@xxxxxxxxxxxxx with the word unsubscribe in the subject line.