As far as I know, creating a video iin Vegas is like making a cake.

You get a big bowl and you put all your ingredients in. You mix them up and pour the batter into floured cake pans. That corresponds to putting everything on the timeline and making you edits, FX, and so on.

Now you put the cakes pans in the oven at 350 degrees, unless they are glass pans, which would be 325. That corresponds to rendering and choosing your rendering templates.

Next you frost your cake. (I like lots of high quality frosting so that it is overflowing with frosting, but I digress.) That corresponds to authoring a DVD or uploading to YouTube, Vimeo, etc. This is the step where you are getting it ready for presentation to the audience.

So I think getting your product out to your viewers without rendering would be like giving everyone at the birthday party a bowl with the cake batter.

I've heard statements elsewhere about how there is no "rendering" in Edius, but I haven't researched it and I can't get my mind around it knowing the Vegss paradigm.

Regards,

Kimberly

Thanks. The computer I have is very slow. For example a 45 minute video usually takes many hours (overnight). So my boss' concern is that we won't have time to render. I've let him know that there's no way around it. I have a new box coming, 6-core with 64gb ram and SSD rendering drive...hopefully it arrives in time and speeds things up.

Thanks for your confirmation.

Shawn

The consensus I see around here is that rendering to a conventional drive is preferable. Disk throughput is never a problem with video rendering; it's all about the CPU (and some would say the GPU).

The new PC should definitely solve your render time issues.

Meanwhile, are you in a metropolitan area?

You should be able to find someone to do basic editing on a 15 minute video for a very reasonable fee...

"Meanwhile, are you in a metropolitan area?

You should be able to find someone to do basic editing on a 15 minute video for a very reasonable fee..."

Why does it have to be someone local? Ever hear of the Internet?

If you are stuck with a s-l-o-w machine, you could run some testing/screening renders at low resolution and low quality. That would take less time than a production quality render.. You will probably render, watch and critique, makes changes, re-render, repeat. The low res render might save some time on that piece.

try avidemux,
it will do the job fast, without recompressing

One thing to think about, if you haven't already, is optimizing render times by making sure that your hard drives function efficiently.

Ideally, you'll have Sony Vegas on one drive, your source media on another drive, and your destination for rendered video on a third drive. This way, each drive will be free to read and/or write without interruption.

There are other things that can be done to optimize render efficiency, but these are, in my understanding, the most important.

This consideration is useful for any machine, slow or fast.

j

Drive throughput, or "efficiency," is almost never a rendering bottleneck.
Unless you're doing 4K over SDI, a single EIDE should do fine.
One should spend their $ on a CPU.

Thinking along the lines of thrashing being the bigger issue, but depends on situation (edit complexity, source data rate, etc.).

My rules include:
-Keeping Windows on one drive with programs (I used to put programs on separate drive but lazy these days & as Windows developed tricky things arose about programs on separate hard drives),
-Puting source media on another drive,
-Setting target render location to a third drive.
-I also regularly defragment my drives (via Windows).

My data drives' throughput seem to top out at about 22 mbps (varies some). If I have stacked tracks of 17mps data-rate media in a project, all on the same source drive, I've already exceeded the drives read rate. In such a scenario I'd consider separating the stacked source.

Requiring a hard-drive to simultaneously write via one process and read via another process (known as thrashing) seriously affects throughput of both operations.

Am I stuck in a time warp (i.e. inappropriately hanging on to obsolete considerations)?
j

I only realized the other day while using a disk rescue tool that SATA has master and slave drives per controller, as did parallel IDE. For best performance you would want the source and destination drives to be on different controllers. Straining at gnats?

I still do things the old fashioned way as above. Vegas on the system drive. Source files on a dedicated drive, and render to another dedicated drive. That's just "best practise". What worked in the past on slooooow machines still works on today's faster bells 'n' whistles' machines. Why wouldn't it?

Those are all dandy "rules", except the last one.
Defragging media storage drives, ever, is a bad idea in my opinion. I won't debate the reasons, but they are real. I've even lost archives in the past.

That being said, none of that is going to improve net throughput in an editing environment, in any significant fashion. Whether one configuration or another reduces the chance of a drive failure is a subject of debate (but not here!).

"

Well, you've just defused your own argument. A single, 10,000 rpm EIDE (PATA) drive may sustain 27 MBps (I'm sure you meant megabytes, not megabits).
A single cheap SATA should sustain 100 MBps or maybe a lot better, under various conditions. Render demands vary a lot, otoh, but are generally expressed in megabits per second, not megabytes.

As I mentioned, when you're working with uncompressed 4K . . .

"-I also regularly defragment my drives (via Windows)."

Jerald,

at a rough guess based on testing i've done, you probably waste somewhere around 1,000,000+ times as much time with the defrag as you save. Drive controllers are so intelligent and fast now that even a nearly totally scrambled drive can perform almost as fast as a defragged drive. Also consider that in real world situations the drive often must do things other than continuously read/write one particular file, so often the head is going to get bounced around anyway. Even in an ideal situation where you're reading one file from one drive and copying it to another, the time savings from being defragged is so small as to be unnoticeable.

The downside is that defragging is a HUGELY ENORMOUS amount of wear and tear on the drive, runs it continuously for long periods of time potentially overheating it, shortens drive life, and also puts your data in jeopardy by destructively re-writing it for no good reason. It's also time spent that could have been better used for other tasks.

As far as folks claiming that defragging allows their system to run more reliably with less crashes, i just have no idea where that idea comes from.

If you really feel you must defrag, then the best way to do it is to always keep an empty drive in your computer. Format the empty, then copy the files from the drive you want to defrag over to the empty drive. The files are written sequentially so they are automatically unfragmented. The process is way faster than running a defrag, doesn't move the heads anywhere near as much or take as long so it's easier on the drive, and it doesn't destroy the original data in the process. You get a chance to make sure everything is fine and dandy before you then wipe the original drive to make it your new empty one. Still though, even this is really wasted effort.

Not to mention media data may actually transfer a bit slower on a freshly-defragged drive.

Can you guys who are against defragging point me to a source that backs your claims? Everything I read online still recommends defragging harddrives. In fact, most people say you don't need to defrag your harddrive in Windows 7 because Win 7 will do it for you automatically.

I think the wear and tear argument is exaggerated. The harddrive head is constantly moving and spinning anyway, so defragging makes it move a bit more, but not HUGELY ENORMOUS. You say it shortens drive life, what is the basis for this information, facts or tests?

You say it slows down data retrieval, everything I see online says it speeds it up. Can you back up this claim?

Because of the way media data is indexed, defragging seems to do no real good, and data is susceptible to corruption (it's happened to me).

Whether it speeds up or slows down actual transfer is really a red herring, because the difference is but a drop in the ocean.

And yes, I will seek out those articles I read maybe a decade ago. They pertain to media data, not Office files, for instance.

I do remember the argument that when capturing video to a harddrive in real time, it is best not to defrag. This makes sense to me because the data is captured to the harddrive in the most efficient way in real time. But since few people are capturing in real time anymore, then this argument does not seem valid.

I can find MANY websites that talk about the upside of defragging and very few that give you downsides. The most common upside is that data retrieval is faster and more efficient after defragmentation.

Yes, and since the drive throughput capability is a magnitude of several times any actual render demands most of us are likely to create, it makes not a whit of difference. With the spectre of possible data loss, it's something I don't do. Much safer to transfer to a fresh drive, as Kelly suggests.

Since this is clearly a side discussion, why don't we continue in a separate thread? See you there. [Emphasis added]

The "faster and more efficient" argument is obsolete, left over from 15 to 20 years ago, when drives were slow and small enough that it did make a difference. Almost everyone repeating the claim now is merely quoting the obsolete argument.

The 'don't defrag when capturing real-time' is a complete red-herring. Aside from the fact that defragging doesn't help, the statement that the file is written linearly is bogus. The file is written to fill the space that is available, which may be in fragments already due to previous file deletions. I'm not saying that defragging would help in this situation, but not defragging certainly doesn't have anything to do with it either.

Really, hard drive speed has increased much faster than the need for size and speed of moving data. Generally speaking, hard drives are about the last place a bottleneck occurs in any situation other than a straight file copy. And, straight file copies are so much faster than any other operation you might perform, so who cares? I remember ages ago backing up PCs are work so we could wipe and reinstall them. The usual method was to mount a spare hard drive to the 2nd IDE port and do a copy. 30MB could take 3 to 5 hours. Yesterday i copied 29GB from one PC, over wifi network to my laptop, onto an external USB drive. Total time: about 12 minutes. That's about 1,000 times as much data moved in about 1/20 the time, and neither of the drives in question has ever been defragged.

Why worry about eking out a fraction of a percentage point over local connections?

Chienworks, most of the websites that advocate defragging that I found are within the last 3 years, not 15 to 20. Even MS states that drives work more efficiently when defragmenting is run.

But I also noticed again, that you are not providing test or data to back up your claims. Granted, drives are faster than they used to be. But I still don't see evidence that defragging a drive causes the problems you claim, but I see lots of websites that claim that head thrashing is actually less after defragging.

You were concerned that people who made the claim that their computer ran better after defragging could not back it up, so I make the same challenge to your claims. Back them up.


" As far as folks claiming that defragging allows their system to run more reliably with less crashes, i just have no idea where that idea comes from."

(now I am not meaning disrespect to either Chienworks or Musicvid, I respect their expertise in video and their input on this site highly. I also like a good discussion)

New thread, guys?