#MAX & PAR – Testing
7 messages in this thread
Greg Pyros & I have been planning something that may be of interest to
the rest of the forumites around here. We both deal in a "non-linear"
animation recording system. Greg handles (and helped design pieces
of) the DPS PAR system. I "play" with Matrox toys, and sell a few,
and have had a little bit of input, on the Matrox MAX.
Both are shipping, maybe, this week, or next. (We Hope <G>) We want to
do a FAIR, PLANNED, and reasonably controlled comparison. This would
take the form of:
A. Written specification comparison
B. Operation and feature comparison
C. Quality of final result (tapes & signals)
The purpose of this communication is to open ourselves up for abuse and
hopefully arrive at a methodlogy for each of those three sections. We therefore
solicit recommendations. From Artists – From Engineers – From Users.
We are thinking of requesting projects or TGA's to lay to the same tape (Beta –
PVW 2800) in some form and combination.
Results by NAB, (maybe Matrox & DPS will buy the drinks)
A bit of a disclaimer:
I, James A. Pope, am the President of Capital Software Development,
which is part of PROSeries Computers, Ltd. of Aurora, CO (DENVER) I
am not being paid by Matrox to do this, it was my own idiotic idea, I
even have to buy the equipment.
I can not speak for Greg Pyros, but he is in the same boat I am.
This is a friendly, "Battle of the Non-Linear Recorders", the rules
will be set by Greg and myself, you can be the judges. The rights to
the results will belong to someone, but Greg & I haven't talked about
that. Maybe we will co-author an article for 3DS Artist or just post
it here.
Feed back is welcome, justified criticism will be at least read, un-justified
remarks will be ignored.
LET THE BATTLE BEGIN
Jim Pope
PROSeries, Denver
Sounds like an interesting project but since you are doing the
shootout with two technologies that are order of magnitudes apart
price wise why not include the Abekas DDR technology as well. It's
about 3X more expensive as the MAX as the MAX is 3X more expensive
than the PAR. With three points on the curve we may be able to show
a relationship the to "inrcrease performance from 90% to 95% one must
triple the cost of his investment" type of proof.
Also, you will need a benchmark to establish the 100% range. I would
suggest the ATVISTA frame accurate to the same recorder.
The rendering should include hirez detailed texturemaps and solid
colors as well as single hue vertical and horizontal gradients. For
chroma integrity use reds against white, cyans against white, and
greens against white.
I would not suggest D1 because of its supperior chroma saturation
curves that might taint the position of the Abekas DDR unless
recording to the D1 in the analog. If you come up with a test I know
of one facility that could do the tests.: Paul LInd, at Anim8, he
gets every new toy and will be most eager to pit one against the
other.
I'm not sure what you hope to prove with this test Jim. In terms of
compression they both use the same chip and technology. As the PAR is
an IDE you won't be able to tandem a couple of hard drives. In terms
of final output one is at non broadcast format and the other is. I
would buy the PAR tomorrow if the frame was 720×486 because its the
least expensive of the two. I might even be able in some
circumstances to use it for final output. My final format will be DDR
to Beta SP and nothing is going to beat it. At least nothing
currently available. So I think the whole idea is rather a moot
point. Those who have to have 720×486 are going to buy the MAX,
everyone else is going to buy the PAR. Why? because the difference
is quality is going to be so neglible as to be irrelevant except for
the most demanding Broadcast environments. All this test is going to
do is prove that the PAR is a heck of a value, and Matrox is going to
have to rethink their pricing. Which they probably won't because those
that need 720×486 are going to have to buy their product, and as it
is still cheaper than a UVW-1800 plenty will. So have fun but
personally the subject is irrelevant as far as I'm concerned. One
thought though, just for realistic comparison, see how well the MAX
and PAR compare after 10 generation dubs. That's what's going to
count in Broadcast. Somehow I suspect you both will come out winners
on this, just a hunch. <G>
-JE
How about adding rotoscoping to the testing parameters. Greg and I are doing
that already!
Chris R.
Chris – Its not my game, but I don't see why rotoscoping shouldn't be
included. The MAX with the 2MegDVE has a built in TBC also. I would
say everything you can compare it with should be checked. And as I've
said before the PAR is a very good value and I expect will compete
very favorably, especially for the price. Unfortuneately, its doesn't
fit my requirements. Wish you guys the best on your "Battle of the
PAR's."!
-JE
Chris,
<<rotoscoping… in the test>>
since many individuals only have high-8 cameras,
i'm looking forward to seeing a hi-8 image rotoscoped
with the par/max uploaded in the forum.
jonas[adesk]
Jim,
I like your idea in theory, but in practice I think it's nothing more than an
academic exercise. The price difference is so huge that the MAX is just
unlikely to be of interest to anyone unless:
1) they really needed a framebuffer anyway.
2) they want to build an off-line, non-linear editing system.