#PAR Excellence…
24 messages in this thread
Well…. after many trials and tribulations with the Seagate 3655A,
literally spending 40 hours plus, tearing apart my system checking
for conflicts of every nature imaginable, trying block sizes or
various settings and generally wondering if I'd taken that final step
over the edge of sanity, (a real concern for those of us living on
the edge <g>), I came to the conclusion that the hard disk was not
giving me the performance and quality I minimally expected from the
PAR. I then contacted Greg P. who was very understanding. He made
several calls including one to DPS upon which we resovled to replace
the drive and rather than take any chances, exchange it for the
Micropolis 2210A which was specifically designed for Video
Multi-media applications. Its also heavier duty and has a built in
shock mount. As before, I took my time installing it, and built the
system from the motherboard level up. Typically, all else being
equal, the PAR could be installed in less than 20 mins. But like many
of todays PC based systems, there's alot of potential for conflict.
So I read the manual and take my time.
Anyway, the long and the short of it is that the image quality is far
better than what I saw at Siggraph or on my friends Amiga based
system. I attribute this to the superior performance of the
Micropolis. One animation I've been using as a test, I was unable to
get the PAR to playback with the Seagate without break-up at a Block
size of 200. The same animation I was able to compress on the
Micropolis with a Block size of 240 and could probably have gone to
250. Suffice it to say that at 240 the image quality in playback is
significantly better than the best I'd hoped for.
On the framebuffer output side, for those technically inclined it has
a video bandwith of -3 db > 5.5 Mhz which is roughly comparable to
SVHS. The higher the bandwidth the more detail and better resolution
potential. In comparison with Beta SP which has a 5.7 sync tip
(lowest video base level) and 7.7 White Peak level. These values are
important in determining output resolution capability, because the
wider the bandwidth, the higher the Horizontal resolution potential.
But as an additional comparison, the Sony CRV has a bandwidth of 4.5
Mhz and is considered Betacam quality as opposed to Beta SP with a
signal to noise ratio of 48 db as opposed to the Par's which is 60 db
(better) and considered professionally acceptable. The PAR falls
between the two in terms of its framebuffer output quality. The
detail in the picture quality will depend on what block size and Q
factor you use, as much as it will the framebuffers potential output
quality. I think the PAR is a professionally acceptable output
device, with better output potential than SVHS because of the Beta SP
component output. Whether it will meet the expectations of more
discriminating clients will more than likely depend on the quality
level they are accustom to. I would much prefer the PAR on Beta SP
tape than on SVHS as a final medium. But don't settle for less than
the Micropolis as your compression drive. The difference between it
and the Seagate, is what puts the PAR in the professional quality
category IMO. And that's my final word on the subject. <g>
After the Micropolis drives came out, we never recomend the seagates anymore.
David,
>> After the Micropolis drives came out, we never recomend the seagates
anymore. <<
Be that as it may, but in the PC version of the manual with a Release
Version of 1.1, Feb 1994 the docs list both the Seagates and the
Micropolis. Why I have no idea, because the Seagates, while recently
upgraded are not in the least up to it.
But as I have your attention, <g> I also have a request with regard to
the GPI trigger. Could you enable it as a trigger for Grabbing frames
as well? Being able to use the GPI for playback is fine but for frame
grabs would be really excellent. There would not need to be any
delay.
John, I have been doing some experimenting here tonight with the GPI
trigger and playback for frame grabs. While, I find it would be
interesting to have this capability the only way I would be able to
get it to work ( produce the required latch pulse on playback ) would
be to set up a normal insert edit with edit VCR and playback VCR
doing a normal video insert. Then, set the edit in point on the
recorder and the playback VCR and begin the edit. A GPI trigger pulse
is NOT sent by the mere actuating of the play button on the VCR
although I could probably build a simple circuit to accomplish this.
The difference is that I would feed the video signal from the
playback VCR to the PAR/TBC IV rather than to the edit VCR. IF the
PAR were capable of triggering the record mode via a GPI latch pulse
we would be in business. However, I would not bother to use this
feature because to generate the GPI pulse from my conventional edit
system would require an overly complicated playback setup. Your EVO
system may be different. As you know I've been doing a bunch of
frame grabbing over the past week and find that just selecting E/E
for video pass through and when the right video comes along just
start the record process. To get the exact starting frame in the
grab just use the split file function and delete that that you don't
need. simple!
Have you hooked an O scope to your GPI output on the EVO to see what type of
signal and when it occurs?
Our GPI triggering devices are completely different. But I see no
reason why it wouldn't work for you just as well. Typically GPI
triggering works in preview mode as well. So you wouldn't have to do
an insert edit. On my EVO however its a bit different. It outputs a
GPI signal at the in-point on playback, which makes it ideal for
frame grabs. And its a piece of cake to write a little program to do
the preroll, Rec play and GPI trigger. So for me it would work very
well to have the GPI trigger frame grabs. This deck was very well
thought out in terms of the needs of animator. Do you remember my
telling you about being able to play back field recorded video and
not see the field shimmer. I've learned that it is a result of an
interlacing technique which they appropriately call "dummy
intelacing" and the beauty is that it looks just like a standard
still image. You can't tell the difference. So for even fast mostion
video like Jim L. was talking about it will freeze an image with no
shimmer. If you've got another name for it, cut me some slack O.K.?
<G>
Sounds like a good idea.
>> Sounds like a good idea. <<
Seems like it would be fairly easy to implement and would definitely be an
added benefit.
Thanks in advance,
-JE
>>And that's my final word on the subject. <g>
with that <g> at the end I know you are going to respond with more. I just know
it! <G>
BTW do you want Rush's book now or should I wait until you do your
first upgrade to the next level of hard drive development after the
Mic 2210A? Bet you can't wait to see the output with a drive that'll
do 450 BLF. Biggest G all day.
>> with that <g> at the end I know you are going to respond with more. <<
You know me pretty well, but I seriously don't think there's anything
more I could add. And anyway, aren't you the spokesman for this
product? <g> I'm just here to keep you honest. <BG>
Now let me see… what was that I did to deserve Rush's book? Oh yes,
I remember. "See I told you so…" <g> BUT my friend that was with
regard to the PAR replacing frame by frame! And that hasn't happened
yet. So you can't send it. 🙂 And by the time it might even be true,
the book will be dated and the Rush will go the way of all wind bag
Conservatives. <BG> Now if you'd said the PAR would replace Hi-8 or
SVHS frame by frame, I think you'd have a strong case, but you
didn't. <BG> So as much as I hate not getting the book, we'll just
have to wait and see.
But I will grant you your glory, it is a serious production tool and
better than I expected. You've got to give Greg some credit on this
too, after all its seems that his efforts and those of Gus made it
possible on the PC platform. And the timing on the Micropolis drive
couldn't have been better.
Now if they'd just implement that GPI for frame grabbing. <g> 0 wait
state please. 🙂
-JE
>> You've got to give Greg some credit on this too, after all its seems that
>> his efforts and those of Gus made it possible on the PC platform.
I wouldn't say we made it "possible" <g>, but thanks for the credit!
Greg Pyros
>> aren't you the spokesman for this product? <g> I'm just here to keep you
honest. <BG>
No! I'm not. If the PAR didn't perform I would have just as quickly come down
on the other side on this device. But, I was pretty sure because I did a
pretty thorough job of checking it out first and as you said with people like
Greg and Gus in the development of this for the PC we were in pretty safe
hands.
>> Now if you'd said the PAR would replace Hi-8 or SVHS frame by
frame, I think you'd have a strong case, but you didn't. <BG>
Actually I never said the "PAR" was going to replace anything. What I
did say was that real time animation recording was going to replace
frame by frame. I never gave the PAR credit for anything other than
being first to replace FBF at the lower end and I did say that the
higher end is already being done by the Abekas. Of course the record
on this in in my thread archives is clear and it has been said that
FBF will never die and will always be the main way we record
animation. Sure, I agree with this just like I agreed with an old
professor in college that told me to "learn the slide rule well
because you'll be using one the rest of your enginneering career. "
What a joke!
As far as the GPI trigger for playback goes: I didn't mean to imply
that I actually had to do the insert edit too it's just that in order
for the trigger to work "I" would have to set it up that way. Your
deck is more suited for this purpose. To use it here would be just
an inconvenience requiring too much setup. I do like your 0 wait for
an option as well. I could see this as necessary in some cases. Such
as Gregs no preroll LDR.
Don,
I think other people interpretted your comments as saying "frame by frame"
would be dead in a year as a result of the PAR. And your comments to Alan I.
indicated as much.
Neither I or Paul would disagree that real time animation recording is going to
replace frame by frame at some point. As a matter of fact, I expect non-linear
editing to replace conventional editing, with similar technology, but its been
a long time in the makeing.
Anyway, I don't want to belabor the point. Suffice it to say, you knew a good
thing when you saw it, and you've probably taken more full advantage of the PAR
than anybody here on the forum. Actually, it sounds like you've already made it
pay for itself. So credit where credit is due. But save the Rush book, Huh? 🙂
Regards,
-JE
>> a result of the PAR
I reread that message to John T. and subsequently to Alan and I think I now see
how you misunderstood all along what I was claiming. Actually I never thought
it was such a unique observation of changing technology. I can understand how
I was mistook to claim that it would be the PAR that would achieve this. I
originally had my doubts about this PAR, but now about 1 month AP I might be
willing to go along with what everybody claimed I said, or until some
competitor comes along and takes some of the glory from DPS. In the time BP
the only viable real time Recording was the highend Abekas. Now that the low
end is covered? what's left? I don't think it will be the MAX. Possibly
something from Truevision but I really think Abekas will produce an Abekas lite
or ultralite to cover the midrange competing with MAX.
BTW, I put the Diaquest up for sale and didn't even get one response. Not even
a curiosity seeker. Last July I had an offer from someone to buy the setup.
How things change! How much FBF have you done AP? What really surprises me is
how little I'm using the targa+ now. I only use it for mat ed. rendering.
Now Don as you keep belaboring the point in your message to Alan you said you
couldn't wait (within the time frame of a year) to be able to say "See I told
you so", about frame by frame being dead which coincided with the introduction
of the PAR. 🙂 But once again we are very much in agreement now that you've
clarified your position.
>> Now that the low end is covered? what's left? I don't think it will be the
MAX. <<
I agree it probably won't be the MAX, it Video bandwidth is just marginally
better than the PAR 5.5 vs 5.7 but if that's 5.7 at +/- .5 db its a bit better.
Even so Beta SP can go to 7.4 with a 220% white level headroom. I don't know
what the Targa 2000's video bandwidth is but I strongly suspect it won't be in
that range.
The only systems that are going to be able to match that kind or performance
are Abekas with Exabyte as the Illuminator Pro is responsible for the MAX's 5.7
video bandwidth. Which means even frame by frame isn't going to be any better
than the bandwidth of the framebuffer. I haven't seen the specs on the Targa
2000 so I don't know how it compares, but I'd be really suprised if it was more
than 5.7
So an Abekas lite seems like a good bet. NAB should have some interesting
products in this regard, if not prices.
Putting the Diaquest up for sale and not getting a response is probably not all
that suprising. After all you've been really slamming frame by frame. <g> I
didn't sell my Targa 32+ when I listed it at $800, so I've decided to keep it
and I'm using it just as much as I did before. Its nice for doing tests and
actually I'm glad I didn't sell it. FWIW
-JE
The bandwidth issue is a curious one and I'm glad your using it as a measure of
quality of the output rather than the nebulous "lines" I agree with your
statements 100% now.
I am curious about your still using the targa+ for tests. With the DPSPAR.pxp
I just load that up and then render a still to the PAR drive with no display
and I get the test render at full or center screen half resolution. I found
two disadvantages to this over the targa+. It is slower due to the 50% higher
hor. resolution rendering ( unless rendering to a box center screen at 376×240
and then it's faster) and some time one must go into the PAR drive and delete
all those test stills. <g> $800 seems like a good deal would you be willing
to sell it for that price to someone here? I know of a guy that was looking
for one. He was going to buy Paul Lind's 64 for $1000 but Paul didn't want to
talk to the guy when I found him a buyer. That's when he explained about his
new privacy phone policy. I think the fact that this guy is a college student
scared Paul off.
I think if I still have the DQ and targa+ by the end of the year I'll donate it
as a package to a local college and take a tax deduction.
Don,
>> I'm glad your using it as a measure of quality … rather than the nebulous
"lines"… <<
Well as you know there's a direct correlation as higher bandwidth has the
potential to generate higher resolution. Most designers will only go to the
expense of higher bandwidth for higher resolution so its a fairly safe
comparison.
I use the Targa 32+ for two reasons. One is the psychological aspect of seeing
the image full screen. The other is the way my system is using a file server
for rendering images; DPSPAR won't work. My three workstations are full, slots
and hard-drives, which is why I went the server route. I don't want to buy
another MB until I'm confident about the Pentiums, and frankly I can use the
extra slots. But its continually evolving process.
I think Paul sold the 64 which is why he didn't want to talk to the guy, but
I'm not sure why he wouldn't have told you this. As I recall Paul said it had
sold very fast. FWIW
I think donating the DQ and Targa + is a great idea! And a good tax deduction
as well. Better than selling it, which my accountant frowns on. Messes up the
depreciation schedule or something. <g>
-JE
John & Don,
I must've forgot about Don's friend. Sorry.
I actually still have more Targa+ stuff and maybe a Vista or two to sell in the
next few months, plus a couple of Diaquest units and maybe a PVW or two.
We're slowly moving to SGI. I'll still have at least 5 copies of Studio running
with Vista's but beyond that, we're going to NFS everything and use the Abekas
and Galileo units to move frames.
Paul
Paul,
>> We're slowly moving to SGI. <<
As much as I hate to hear it, I do understand. Can't say I have a need for the
other stuff ie: PVW or Vista but I'll keep it in mind. Don's friend may still
need a Targa.
How do you like the Galileo? Is is comparable to the Abekas or would you rather
buy another Abekas should the need arise?
-JE
Paul,
Keep me informed when you have the equipment for sale. I have one client who
is looking for a Targa+ board, so far they have been bidding low but they may
change and we could always use a PVW depending on the cash flow (+/-). And a
VISTA is always on my list.
Thanks
Fred
"I originally had my doubts about this PAR"
I guess this is where we get to tell the story when I almost strangled you at
ImageWorld?
🙂
>>I guess this is where we get to tell the story when I almost strangled you at
ImageWorld?
🙂
Yes, I know I was getting you pretty upset that day with the starfield glitch.
It's just that your answers wern't satisfying me. When I want something really
bad I do my best to find everything wrong I can with what I see and that was
it. I did the same at the Matrox booth and the Bandit booth. I had similar
experiences there too. Later I came home and through use of motion blur I
was able to recreate that double starfield defect in my frame by frame system.
Consequently, I was convinced it was not the PAR but in how that particular
animation was rendered. Accepting donated animations is a double edged sword,
you know.
Actually, you can check this with GregP. and John Ellis, in your absence during
December and January, I think I did quite a bit to dispell much of the trash
that was being put out about the PAR like its "terrible jpeg video quality"
Some actually accused me of being on your payroll. <G> Your product is doing
everything you claim and then some. Other than minor bugs in the software,
which we all have learned to expect in this business, everybody seems to be
very pleased, and pleasantly surprised at the PAR quality. Keep up your
presence here on the forum and try not to strangle any of us when we get rather
obnoxious in the interest of solving problems or pointing them out.
>Accepting donated animations is a double edged sword, you know.
How well I know! But to all out there, we are in need of additional animations
for NAB and Sigraph. If any one has rendered frames that we could use, please
contact me. We can accept most tape backup formats and programs.
David White
Computer Products Manager
Digital Processings Systems
Thanks for the info on the PAR and knowing in advance to pick up the
Micropolis drive will save me some serious aggravation. This assumes
I can get my company to fork over the bucks to buy the card in the
first place.
Bob,
Best of luck with your company advancing the money. I think if you could get
the PAR demo-ed they'd be on it in a heart beat! Especially excellent if you
need to multiple first generation output.
Keep us posted. 🙂
-JE