comp.lang.idl-pvwave archive: archive » Testers needed for TV benchmark

Home » Public Forums » archive » Testers needed for TV benchmark

Show: Today's Messages :: Show Polls :: Message Navigator
E-mail to friend

Testers needed for TV benchmark [message #26144]

Wed, 08 August 2001 10:24

billb
Messages: 5
Registered: August 2001

Junior Member

Hi all,

I've been working at trying to get the fastest TV update rate possible
and am a little discouraged by my results so far. I developed a short
benchmark program and am curious what results might be obtained by
other machines. I have a P3-550 and obtained the following:

TV-Device Copy Test: Array Size: 512 13.520000 seconds
=> Frames/sec: 3.6982249 Frame time: 0.27040000
Video Xfer rate: 2908402.4 bytes/sec

TV Direct Test: Array Size: 512 13.290000 seconds =>
Frames/sec: 3.7622272 Frame time: 0.26580000
Video Xfer rate: 2958735.9 bytes/sec

If your time permits, I would appreciate knowing anyone else's results
and a brief mention of the computer used. My benchmarks were
identical for IDL 5.2 and 5.3.1. Thanks in advance.

Bill B.

============================================================ =================

; Benchmark test for window updating

PRO WIN_TEST, GROUP=Group

set_plot, 'win'
DEVICE, DECOMPOSED = 0 ; New for IDL 5.2

sz = 512

WINDOW, /FREE, /PIXMAP, XSIZE = sz, YSIZE = sz
pixmap0_id = !D.WINDOW
WINDOW, /FREE, XSIZE = sz, YSIZE = sz
pixmap1_id = !D.WINDOW

R = FINDGEN(256)
G = R
B = R

TVLCT, R, G, B ; Load these vectors into the color table:

s = 0 ; seed
data = (randomu(s, sz, sz, 3) * 255) ; create some initial data

; TV to pixmap and copy test -----------------------------

T0 = SYSTIME(1)
FOR I = 0, 49 DO BEGIN
WSET, pixmap0_id

;data = (randomu(s, sz, sz, 3) * 255)
;s = s + 1

TV, data, true = 3

WSET, pixmap1_id
DEVICE, COPY = [0, 0, sz-1, sz-1, 0, 0, pixmap0_id]
ENDFOR

T1 = SYSTIME(1)
fps = 50.0 / (t1 - t0)
print, 'TV-Device Copy Test: Array Size: ', sz, ' ',t1 - t0, '
seconds', ' => Frames/sec: ', fps, ' Frame time:', 1.0/fps
print, ' Video Xfer rate: ', DOUBLE(sz) * DOUBLE(sz) * 3D * fps, '
bytes/sec'
print

; TV direct test ------------------------------------------
T0 = SYSTIME(1)
FOR I = 0, 49 DO BEGIN
WSET, pixmap1_id

;data = (randomu(s, sz, sz, 3) * 255)
;s = s + 1

TV, data, true = 3

ENDFOR

T1 = SYSTIME(1)
fps = 50.0 / (t1 - t0)
print, 'TV Direct Test: Array Size: ', sz, ' ',t1 - t0, ' seconds', '
=> Frames/sec: ', fps, ' Frame time:', 1.0/fps
print, ' Video Xfer rate: ', DOUBLE(sz) * DOUBLE(sz) * 3D * fps, '
bytes/sec'

WDELETE, pixmap0_id
WDELETE, pixmap1_id
END

Report message to a moderator

Re: Testers needed for TV benchmark [message #26181 is a reply to message #26144]

Fri, 10 August 2001 09:11

Dennis Boccippio
Messages: 23
Registered: July 2000

Junior Member

IDL 5.3, MacOS 9.1, Powerbook G3/300:

TV-Device Copy Test: Array Size: 512 6.1833333 seconds
=> Frames/sec:
8.0862534 Frame time: 0.12366667
Video Xfer rate: 6359288.5bytes/sec
TV Direct Test: Array Size: 512 5.2666667 seconds
=> Frames/sec:
9.4936709 Frame time: 0.10533333
Video Xfer rate: 7466126.6bytes/sec

In article <269b6343.0108080924.2bd48749@posting.google.com>,
billb@attinet.com (Bill B.) wrote:

> Hi all,
>
> I've been working at trying to get the fastest TV update rate possible
> and am a little discouraged by my results so far. I developed a short
> benchmark program and am curious what results might be obtained by
> other machines. I have a P3-550 and obtained the following:
>
> TV-Device Copy Test: Array Size: 512 13.520000 seconds
> => Frames/sec: 3.6982249 Frame time: 0.27040000
> Video Xfer rate: 2908402.4 bytes/sec
>
> TV Direct Test: Array Size: 512 13.290000 seconds =>
> Frames/sec: 3.7622272 Frame time: 0.26580000
> Video Xfer rate: 2958735.9 bytes/sec
>
> If your time permits, I would appreciate knowing anyone else's results
> and a brief mention of the computer used. My benchmarks were
> identical for IDL 5.2 and 5.3.1. Thanks in advance.
>
> Bill B.
>

Report message to a moderator

Re: Testers needed for TV benchmark [message #26214 is a reply to message #26144]

Thu, 09 August 2001 09:29

david[2]
Messages: 100
Registered: June 2001

Senior Member

Karl Schultz writes:

> Even when the X server implementation is better, that is, the
> off-screen pixmaps are faster, this can cause confusing benchmark
> results since the copy times can vary in hard-to-describe ways.

In my experience, nearly everything about pixmaps
is hard to describe. :-(

Certainly it is hard to explain why a program
works on *your* computer, but not your boss's
computer, which was ordered from the supplier
on the same PO and from the same computer
manufacturer.

That's why I always confine any explanation to
*small* animations and lots of hand-waving.

Cheers,

David

--
David Fanning, Ph.D.
Fanning Software Consulting
Phone: 970-221-0438 E-Mail: davidf@dfanning.com
Coyote's Guide to IDL Programming: http://www.dfanning.com/
Toll-Free IDL Book Orders: 1-888-461-0155

Report message to a moderator

Re: Testers needed for TV benchmark [message #26282 is a reply to message #26144]

Mon, 20 August 2001 06:20

engpcris
Messages: 4
Registered: August 2001

Junior Member

My results:

Config:
Windows 2000, IDL 5.4, Athlon 1.2 Ghz, 512 Mo Ram

TV-Device Copy Test: Array Size: 512 6.2090000seconds =>
Frames/sec: 8.0528266 Frame time: 0.12418000
Video Xfer rate: 6333000.5bytes/sec

TV Direct Test: Array Size: 512 4.9970000 seconds =>
Frames/sec: 10.006004 Frame time: 0.099940000
Video Xfer rate: 7869041.5bytes/sec

Good luck.

Report message to a moderator

Previous Topic:	problem with OPENURL
Next Topic:	using Insight with IDL 5.4

-=] Back to Top [=-

[ Syndicate this forum (XML) ] [

] [

]

Current Time: Wed Oct 08 13:19:15 PDT 2025

Total time taken to generate the page: 0.01190 seconds