#Razor Pro Problems
5 messages in this thread
Jamie,
This is a copy of a previously sent message, If you have already received
this, I am sorry for the duplicate. We are really in need of some help getting
Razor working correctly. if you need more info to help us get the problems
working, please let me know.
—
We are having trouble using Razor Pro w/ our PAR board. When we needed it
most, the output made it unusable, and we had to resort to using just the PAR
software (not very pretty, but worked w/o problems).
Now that the deadline is passed, and we have finished our first major project
with the PAR, we really need to find out why Razor is not performing as
expected.
First, I have spoken with Tech support about some of the problems, but so far,
they have not resolved all issues. The problems are as follows.
1) When using an .ani file as one channel of video, with another channel of
sequential .tga's alpha-channeled over this, the final output to .ani had two
places that where jerky and unuseable. We then tried to re-record the sections
that were screwed-up, but they would not change. To see if this was RazorPro,
we then video posted the bad sections from 3ds, and they worked perfectly.
2) When trying to make a .prj file with both .ani's and .wav's, the video was
jerky in a few spots, and the sound was extremely bad. I spoke with tech
support about this, and tried all the suggestions given by them, to no avail,
as the .wav file was recorded uncompressed, I made sure the settings were OK,
etc. I have also upped the buffer on the par to 1500, etc.
Our main concern is the video, as we have fewer sound projects, and we really
would like to use razor rather than video post, however, we need both to work
as advertised.
Two other things, the stuff composited from sequential .tga's to the par was
slightly darker (especially the alpha channeled transparencies) than the stuff
video posted from 3ds. Is this due to a gamma setting, and is there a way to
set the gamma output in Razor Pro?
Also, while making a series of sequential .tga's with RP, when there was only a
single channel of images with no transition, RP would only save a .tga file
with a size of 0, and no image. Why is this and is there a fix?
Thanks for you help in getting these issues resolved. We are desperately
looking for answers.
Doug Glasgow
Georgia Tech Reasearch Asst.
(404)801-8126
Doug,
Here is a copy of the reply that I sent to you last week:
Tolek,
Very sorry to hear that you are having problems with RAZOR PRO. I
hope that I can help you resolve them.
If you are not already using version 1.0d of RAZOR PRO you will
find it on our BBS – (301) 320 0338.
>1) When using an .ani file as one channel of video, with another channel
>of sequential .tga's alpha-channeled over this, the final output to .ani
>had two places that where jerky and unuseable. We then tried to re-record
>the sections that were screwed-up, but they would not change. To see if
>this was RazorPro, we then video posted the bad sections from 3ds, and they
>worked perfectly.
This sounds as though the compression settings for the PAR are not set
correctly. You can check these by clicking on the change button next to
the PAR ANI compressor name on the final output settings dialog box.
Make sure that the Q factor is 23 and that the block limit is 240.
>2) When trying to make a .prj file with both .ani's and .wav's, the
>video was jerky in a few spots, and the sound was extremely bad. I spoke
>with tech support about this, and tried all the suggestions given by them,
>to no avail, as the .wav file was recorded uncompressed, I made sure the
>settings were OK, etc. I have also upped the buffer on the par to 1500, etc.
For video jerkiness and quality see above.
For audio quality please tell me if you put a single track of audio on the
timeline
then make out a final file which is an audio file with 44.1, stereo, 16 bit
quality
whether the audio is still bad. Please let us know the results of this is.
>Two other things, the stuff composited from sequential .tga's to the par
>was slightly darker (especially the alpha channeled transparencies) than
>the stuff video posted from 3ds. Is this due to a gamma setting, and is
>there a way to set the gamma output in Razor Pro?
Are you comparing like with like here ? Did you try compositing the same
exact images ? Please let us know the results of this.
>Also, while making a series of sequential .tga's with RP, when there was
>only a single channel of images with no transition, RP would only save a
>.tga file with a size of 0, and no image. Why is this and is there a fix?
Are the input and the output file names the same ?
Please post the results of the above tests to :
tech-support@in-sync.com
Jamie Carr
in:sync
Jamie,
Thanks for your response. I will test your suggestions and reply to the e-mail
address. With regards to the following:
>>Two other things, the stuff composited from sequential .tga's to the par
>>was slightly darker (especially the alpha channeled transparencies) than
>>the stuff video posted from 3ds. Is this due to a gamma setting, and is
>>there a way to set the gamma output in Razor Pro?
>Are you comparing like with like here ?X)
Scott:
>> I'm leaning toward the PVR I saw at NAB t632 (X)
Greg:
Thanks for the run down, Greg. The WSD/XL's abilis is powering down the system.
Reset DOr mode so my PAR sscreens could also be viewed on mer with QEMM than with DOS' EMM386. What
version of DOS do yDS64 p, at home (my Micron machine) Icess to another video card to try out, ter the VESA or the e for a given resolution & color depth alth 64 cards. They're using the newest S3 chips (968?) which aren't
supported yet. For now, you'll have to configure for VESA and it will work
fine. Sor 16M colors.
I don't have the Diamond c S3/Hware/OpSys/video
25-May-95 01:42:32
Has anybody had experience running the PAR under Win95? I'vr just tried and
while
Win95 lets it run, Explorer and other file management systems fail to read or
access the PAR's directories or files. Looking at the PAR drive under Win95,
you see your main drive. However, the PAR software runs. Whether it loads or
unloads images, I don't know yet.
Martin Kunert
KMe
Doug,
OK. I have done some research on the gamma stuff and here's
the lowdown.
3D Studio is an image generation program. Razor PRO is not
an image generation program its a compositor and editor. When
images are generated (eg rendered) it is best to do any gamma
corrections at this time, not later in post.
The upshot is you need to be correcting the gamma in 3DS. If the
output you are seeing in Razor PRO is not correct then you are
not giving it the right source images and the gamma needs to be
corrected in the rendering process in 3DS before you put the images
to RP.
Having said that, we can write an effect to correct gamma (I've added
it to the list) but my advice is to correct the TGAs you are rendering
in 3DS before you hand the images to RP. (for additional information
on Gamma correction please see the latest issue of Dr Dobbs)
You mention a problem writing to a series of TGAs. We have many,
many people doing this, so I am sort of at a loss to know what the
problem is here. Please tell me how much RAM you have. What
size are the source images, and what size are the output images ?
What color depth are the TGAs you are generating ? Try loading
the pondduck.avi in the examples directory and writing out a series
of 160×120 TGAs. What happens ?
Jamie Carr
in:sync
Jamie,
Thanks for getting back to me about the gamma problem. I will look into
correcting the gamma issues inside of 3ds. As yet, I have not had time to
write to the tech support people at you other e-mail address, so I appreciate
your responses here.
You said:
> You mention a problem writing to a series of TGAs. We have many,
>many people doing this, so I am sort of at a loss to know what the
> problem is here. Please tell me how much RAM you have. What
>size are the source images, and what size are the output images ?
>What color depth are the TGAs you are generating ? Try loading
> the pondduck.avi in the examples directory and writing out a series
>of 160×120 TGAs. What happens ?"
I have just recently experienced the same problem, but as a new user it could
very well be user error. We have 64megs of RAM, the source images are 752×480
and the output images are 752×480 w/ a color depth of 24bit. They are tga's
being rendered to tga's (usually when I was adding an effect in the middle of a
sequence, and was simply rendering the entire sequence, not simply the effect
sequence). I must stress that the parts that have a transition or effect work
correctly, but it is simply the tga's that do not have any effect or transition
that are created with a file size of 0 bytes. Any clue? I will load the
ponduck.avi and give it a try, but I have not actually seen this problem except
with tga to tga.
Thanks again for you help.
I have just finished a rendering for the CNN Center renovation here in Atlanta,
and it was presented to Ted on Tuesday. I used Razor to composite most every
thing and had only one real problem, though I am not sure what caused it (razor
or 3ds). This animation was only 2.5 minutes long, so I didn't have the same
problems as before. For this animation, I rendered the animation geometry
first, and rendered video streams as alpha channels to be composed on top of
the background animation. For some reason when composited in razor, at two
places, the background repeated itself 2 or three times in a row, causing a
glitch in the motion. The rest of the animation matched perfectly. I went
into razor and cut out the glitch, re-applied the "same" tga sequence I had cut
out to the glitch space, and the scene rendered without the "hiccup", though
there was a one frame hitch at the end of the offending segment. Do you happen
to have any ideas on this? It was real confusing. Thanks for your help.
Doug