View Full Version : Anyone up for a test?

09-14-2003, 03:51 PM
I found a test that Media 100 did to comepare themself with Avid and itīs quite fun to read :) .

It would be interesting to see what VT3 can perform in the same kind of test.
I dont have VT3 myself so my Qis if anyone could do this test and show us the result here on the forum?.

Here is the link http://www.media100.com/realtimecomparison.asp

Realtime doh:rolleyes:


09-15-2003, 03:22 AM
Well the test is of course not fair, just look at who is doing it.
It includes in the comparison a few marketing gadgets and puts a big X on the Avid system for not having them.
like "Intelligent layering Architecture, and Visual Voicing"

As I understand it, T3 doesn't need 10bit as it doesn't work with RGB, it keeps video y YUV
It does work in real time with:
Scale, position, Rotate, crop, color correction, Key and blur
Dunno what they mean by "Motion Alpha" but it does support clips with alpha.
It does not have blend modes, altho I've found a way to get around it in some situations by adjusting the layers color and key settings.
It does however have: 3d rotation, smooth borders, multy-colored borders, transparent corners, shadows and those lovable motion splines.

The number of layers you can do with T3 depends on the speed of your processor(s) and disks, 7 is the highest layer number I've heard, but as machines progress so will this number. So the test with the change of news anchor would depend on the host machine.

T3 renders FASTER than realtime in some situations. (like rendering as DV on a dual Xeon)

T3 also includes top notch 2d and 3d software and a live switcher. (even without the break out box you can switch between several computer sources, DV cameras, component, and composite inputs)

I bet both the Avid and Media100 are nice systems, at that price they better be!

Of course to really do the test we would need the same footage =)

09-15-2003, 06:40 AM
Originally posted by Danner
Of course to really do the test we would need the same footage Maybe not, since uncompressed footage places pretty much the same throughput demands on a system regardless of it's content. There is of course, the difference of 8bit vs. 10bit, but without getting into that argument again, this test seems to have ignored it so I suppose we can too...

Anyway, always up for a little challenge with my morning coffee, I took a quick look at this...

On my system (dual 2.8 Xeon with a 'paltry' gig of RAM), adding the seventh layer (an animated lower third, for which I simply scaled a Digital Juice anim, with edges softened and alpha reduced in VT-Edit) forced me to turn on background rendering. Until then, I had it turned off, which is my normal operating mode.

I'm pretty sure I could have gotten away with it completely without rendering if my stripeset was not the older U160 standard. As it is, it took about half the time of the Media 100 to render in the background. Obviously this was not a definitive test, but I think it serves adequately to answer the question. BTW, my greenscreen layers included both RTV and AVI streams.