Yes, there's a setting in Options>Preferences>Video to set the maximum number of rendering threads.

Mike

Thank's Guy's. What type of speed are you getting renedering full HD (1080) on a quad core ?

Thank's Guy's. What type of speed are you getting renedering full HD (1080) on a quad core ?

I don't really know how to answer that because it depends so much on the type of source and destination, not to mention additional processing of transitions, effects, etc.

What I can say is that all four processors work flat out during rendering.

Mike

"Thank's Guy's. What type of speed are you getting renedering full HD (1080) on a quad core ? "

I don't use many transitions or effects, but I do trim most clips.

If I render to SD Video then it takes about twice the length of the video.

If I render to MainConcept (m2t) then it takes a little longer than real time.

However, Options>Preferences>Video do not go higher than 4 threads in VMS HD 10. If you've managed to raise the number, I'd be grateful to know how

Yes, there's a setting in Options>Preferences>Video to set the maximum number of rendering threads.

That setting has nothing to do with the amount of cores used.

So can anyone confirm if they are successfully using ALL 8 threads of a quad-core processor for video rendering ?

It uses a maximum of 4 threads.

Exactly. The maximum number is four

Guy's,

I can confirm that Vegas Movie Studio HD can actually utilise all 8 threads.

Just to clarify - an i7 quad core processor delivers 2 threads per core, so 2x4 = 8.

I have just rendered footage into an m2ts file, and all 8 threads were utilised in the processing.

Yes, all 8 cores were being used .... but that does not mean that the program was using more than 4 threads

"Yes, all 8 cores were being used .... but that does not mean that the program was using more than 4 threads"

I'm interested in the answer to this question as well, and I'm seeing a lot of replies that seem to indicate a misunderstanding of threads and cores.

As has been said before the i7 processor has 4 cores, and each core can process two threads concurrently, so up to 8 threads can be running simultaneously.

Edit: generally the way threads are time multiplexed on the core hardware involves switching from one thread context to another when one of the threads encounters a delay due to cache misses or waiting on an input, etc. So I think its impossible to increase the potential throughput of a single core by allowing it to support two threads per core. However, if the first thread was only able to utilize the core to say 50%, it is theoretically possible for another thread to be swapped in to use the other 50% that the first thread couldn't utilize. So the two threads appear to perform twice the work of a single thread, but their aggregate computational progress can never exceed 100% of the capacity of the core upon which they are running. So the potential increase in throughput by having 8 threads instead of just 4 (one per core) is highly dependent on the nature of the software being executed. If thread A is running at 90% core usage, then thread B can only use the 10% capacity not being used by thread A before the two threads go into contention for the core. Once both threads are in contention for core resources then the efficiency of two threads per core vs. one isn't going to be all that significant.

If VMS can only have 4 threads running at one time, then it can't utilize a i7 to its full potential of 8 threads. That's what we're trying to determine, and also I'd like to know (but doubt if we'll get an answer) if all 8 threads give the equivalent throughput of 8 cores, or somewhere between 4-8 cores. If rendering algorithms are already very efficient then its probably closer to the 4 core performance anyway and i7 cores probably aren't significantly faster than quad core single thread processors.

I've seen people claim the number of VMS threads can be increased beyond 4 by setting a preferences value, but when I tried to do that upon launching the program it would really only accept a maximum value of 4, even though a larger value was temporarily displayed in the preferences menu.

I had read somewhere (some Sony info I'm almost sure) that to use more than 4 cores, required using the 64bit version, and that means Pro.

It would be interesting to see what procexp said about how many actual threads were busy, even if someone snuck in a number larger than 4.

I would expect the most efficient number would be the number of cores you have, and if your proc supports hyper-threading, then cores*2 might be faster. But having 4 times the number of threads does not mean 1/4 of the processing time. And given the amount of memory that video processing goes through, cores*2 might be slower. Memory/disk bandwidth and cache utilization all play large parts also.

This sucks, it's like you're only limited to a maximum 3.75GB of RAM due to a 32bit OS instead of 64bit -- even though you have 8GB RAM installed.