[Dave Haynie]

I love the idea of an SSD… lower power, no effective fragmentation, zero seek time, fast reads (writes can actually be slower than HDDs, aside from the seek-time component), etc. But they’re just too small.

I was cleaning out my “Windows.old” directory… mostly just wasted space, from my old XP system after a Windows 7 upgrade over a year ago. I had many times that 60GB there, including 42GB of Microsoft “Application Data”, 30GB of “Program Files”, and tons of other stuff. Not even counting the “Documents” and “Music” directories, which I don’t really use anyway (most of that stuff goes straight to my RAID). You really have to be able to keep things in much different places than Windows wants by default, for this to work well and not run out of memory.. at least if you’re as “ambitious” in your PC use as I am (my C: drive is a 1.5TB SATA).

In Linux, it would be fairly trivial to use an SSD and put less critical stuff in other places, but use symlinks to make those extra directories show up where you needed them. Or, as an alternative, set up that large hard drive as C:, but use symlinks to local critical system and other apps on a separate Flash drive. Not sure how this would work out on Windows (they did add symlinks to Windows in Vista).

The big issue with SSDs, other than size, is the very limited erase/write cycle life of the drives. If you’re running Windows 7, this is less of a problem, because it detects SSDs and automatically stops defragging and other functions that will have a noticeable effect on drive life. Vista and earlier need special tweaks to prevent these things.

I’m also not comfortable with the swapfile on the flash drive. Yes, modern SSD controllers do wonders with wear leveling, it does make swap much faster, etc. If you have enough RAM, you rarely use the swap file anyway.. but then, no need to make it that fast. If you don’t, you will definitely lower the effective life of that SSD… particularly if your drive uses MLC (Multi-Level Cell) flash, rather than SLC Flash. MLC has about 1/10th (or worse) the write-cycle life of SLC.

-Dave

[Dave Haynie]

YouTube takes most major formats… AVC/MP4, MPEG-2, WMV9, etc. I have uploaded full 1080/24p and 1080/60p video files to YouTube, at bitrates up to about 20Mb/s, and it ate ’em up just fine.

HD rendered to HD will look better than SD rendered to SD, regardless the output format, unless you’re trying to match bitrate (eg, SD at very low bitrates will probably look better than HD rendered to the same very low bitrate). For YouTube, there’s no need to go to a crazy low bitrate for upload, unless you have internet connection issues (and I’m more limited, on satellite, than most folks).

My usual YouTube upload is 1280x720p rendered in AVC at 6Mb/s, which looks great… I have only done a few YouTube projects since they started actually supporting 1080p playback (that’s where the 20Mb/s videos went). Do keep in mind that, no matter what format you provide, they will recompress to their various standard resolutions. So you want to optimize the quality of the upload, knowing that recompression will lower it somewhat.

-Dave

[Dave Haynie]

Just a consideration… did you switch from 32-bit to 64-bit in the process of the upgrade? And what is the input CODEC used for the rendering you’re doing.

Why I’m asking… some of the video CODECs out there have terrible performance in 64-bit mode. No idea why, but I’ve seen that with several, particularly open source stuff like Lagarith (that one was never acceptably fast even in 32-bit, but it got worse).

In any case, if this is your situation, try a more standard video input format, and/or install the 32-bit version and see if things get better. Rendering always takes too long, but I have not experienced any significant difference between Vegas 9 and Vegas 10 (in theory, Sony AVC should be faster in Vegas 10… but the GPU boost isn’t very significant, at least with my current nVidia 8800GT).

-Dave

[Dave Haynie]

I guess everyone’s got their favorite new features. This wasn’t a “knock your socks off” release, but it’s a good one for many purposes.

One big one for me was the inclusion of support for the Open Effects Association plug-in architecture. Vegas had previously only supported DirectShow filters as plug-ins, which greatly limited what a plug-in could accomplish. This fixes the problem, and lets you spend lots more money on new plug-in collections (I bought BCC7 after playing with the trial for a few weeks). For example, the DirectX model didn’t allow a plug-in to process multiple frames without introducing a delay.

Another big one for me: track groups. This was something I had asked for (and received) years back in Acid… when you get past a certainly number of tracks, the “flat” view is insufficient. I had a few music video projects last year with 40+ video tracks… I did not precisely remain sane during those projects.

Some other stuff: multicam support is improved. I don’t shoot anything non-trivial without at least one second camera, so I do multiple camera shoots all the time. Vegas 9’s multicam had potential, but half the time, I found myself doing it “the old way”. Vegas 10 seems to fix this. Audio is once again treated as “first class”… you can have clip-level effects, just as with video… something that always bothered me in the past.

The stabilization is a nice plus. I generally shoot from a tripod or Glidecam, but can’t always. Not so much that I had planned to invest in a top-level stabilization plug-in, but having it built-in, I’ll take it. I’m using it in one current project, and expect to have lots more use for it in the next one.

I don’t always get the upgrade… thanks to Sony’s policies, you don’t pay more by skipping a generation. But it’s what, $140… not free, but if I can count ten things that make my life easier, that’s $14 per feature… not too hard to justify.

-Dave

[Dave Haynie]

People tend to think of “flash == small” and “tape/HDD = large”, but that’s kind of old thinking. A 63 minute DV tape stores 12GB, an 83 minute DV tape stores 16GB… not huge.

Most AVC camcorders record at 24Mb/s or less, at their peak bitrate. That’s under 11GB per hour. So you get almost 3 hours on a 32GB SDHC card. In practice, you may actually get more than 3 hours, particularly shooting a low-motion subject such as a seminar. Most tapeless cameras record in variable bitrate.

And of course, they all offer lower average bitrate modes as well. So for a hit in overall quality, you can get close to six hours on the same memory card.

Some consumer models also support more than 32GB of internal flash memory, or, as you mentioned, hard drives. I’m not a fan of HDD models, as the HDD then becomes the most fragile part of the camera, and they may simply not work at high altitudes. If your camera supports the SDXC card standard, you can currently buy 64GB memory cards, which would give you around six hours of shooting at top quality.

And by “top quality”, I mean high definition. You’re currently shooting in DVCAM on full size tapes (4.5 hours max)… you can definitely get something similar onto a memory card. Maybe HD doesn’t even matter… that’s something to check on the spec sheets — many HD models don’t do SD, but you might be happy with 720p rather than 1080i.

-Dave

[Dave Haynie]

You will re-render.

Firstly, Vegas doesn’t currently do AVC smart rendering. And in Vegas 10, Smart Rendering seems a bit broken anyway… this will hopefully be fixed. Extending to AVC would be better, still, but at present, Sony doesn’t including an AVC CODEC with support for over 20Mb/s… less than many people shoot, certainly less than the 1080/60p rates most so-capable cameras shoot.

Then there’s Blu-Ray. 1080/60p is not a supported Blu-Ray format. If you want to make a Blu-Ray compliant disc (or AVCHD, for that matter), you need to rendering to either 720/60p or 1080/60i.

There is no problem editing in 1080/60p, other than the fast that’s crazy slow on most systems.

-Dave

[Dave Haynie]

Most broadband modems will only communicate with one device at a time, and you may have to power-cycle the modem to get it to communicate with a different PC.

Ideally, you want to add a router… this is a small device that will present as a single Ethernet device to your modem, but allow any device on your network to communicate, using various tricks such as Network Address Translation (NAT). Routers can also offer some degree of firewall protection to the devices you have connected, they offer various controls over kinds of things that happen over the net, etc.

If you don’t want to buy a router, you can sometimes use a PC to do the same job. Microsoft has a built-in “Windows Connection Sharing” function, which lets one device act as a router. You do need multiple Ethernet ports for that (or one Ethernet, one wireless). A better solution, if you have Wireless and Windows 7, is to try Connectify (free download), which has a number of improvements over WCS. The only problem using a PC as a router is that, if you shut it down, all of your other devices lose their connection.

-Dave

[Dave Haynie]

DVD supports 23.976 video… kinda-sorta. You have to render your 24p video as “23.976 + 2-3 pulldown”… you’ll find that option for rendering MPEG-2.

-Dave

[Dave Haynie]

Try re-muxing the Sony output. The MPEG-2 transport stream can be rebuilt without any change to the video or audio sections, but this seems to fix the problems I saw in the Sony-muxed version.

Go to https://www.smlabs.net/tsmuxer_en.html, download the tsMuxeR program. You can drop the Sony output file into this, then select the specifics, such as Blu-Ray or AVCHD output. The program actually creates a formatted AVCHD directory… so this can not only mux your AVC properly, but create the files you need if you’re interested in building an AVCHD DVD (I don’t recall if you said just why you’re making AVCHD output, rather than Blu-Ray or generic MP4).

-Dave

[Dave Haynie]

I don’t understand your comment about “no need for application to support a graphics card.” Of course the OS is where the drivers live to support the app, that’s what I was meaning.

If you don’t have the right language, you can’t express the proper meaning. An application should know as little as possible about the underlying hardware — it talks to an OS-level abstraction. That’s why it works with any graphics card.

Whether the app can output the right quality of resolution would be some code in the app talking to the OS, which is all the tuneable settings in Vegas or is this no longer true in Windows 7?

Vegas works in Windows, not direct-to-hardware. The application, like all applications, sets some specifications when creating a Window, and it gets updates when the user makes modifications to that Window. Windows has a couple of different APIs for speaking to a display; last I heard, Apple uses something fairly similar to Adobe Acrobat to create content in MacOS Windows (perhaps they have something more low-level, like Microsoft’s DirectX… and of course, for some aspects video, they both use the OS’s video APIs… the don’t know a whole lot about the specific video device in use, or even if there is a real video device).

Perhaps my problem is that if I properly calibrate the Windows machine to the monitor (which is what I guess that device will do for me), I’ll get a closer rendition to the Mac output on it. But that’s the point of my post, the Mac just does it without me messing with it…

The Mac doesn’t exactly calibrate itself. But they store the color profile of their monitors in the monitor, and MacOS knows how to pull it over… it’s essentially an ICC profile, hidden in the monitor itself. Apple explains what they do here: https://www.apple.com/pro/color/workflow/basics.html
This is certainly better than nothing, but not as good as doing an actual calibration. Apple calls their color management system “ColorSync”.. here’s another article: https://www.gballard.net/psd/colorsync.html

On Windows, Vista and Windows 7 use the Windows Color System, detailed here: https://www.microsoft.com/whdc/device/display/color/WCS.mspx

Of course, since Apple’s building these in for Macs, unlike other monitor manufacturers, it’s not easy to find the profiles from Apple. But there are collections, like this one, which might help here: https://www.tftcentral.co.uk/articles/icc_profiles.htm . If you find a useful profile for Windows, download it and put it here: C:\WINDOWS\system32\spool\drivers\color.

But again, without actual calibration, you will not have fully corrected color on any monitor. Period. It has nothing at all to do with operating systems, it’s the nature of color on monitors. And printers, and any other output device you use. And the TV in your livingroom (assuming you have a TV, and/or a livingroom).

If there’s something I need to know about settings on the Windows machine, (i.e. what *are* the right settings for us in the project properties dialogue? there is no training material I’ve seen to help me understand when to change those settings!)

There are no “right” settings… it depends on what you’re after. Vegas is a tool, and not the easiest one on the planet to use — the best tools are always more complex than the simpler ones. Read the manual. If you’re still confused, there are plenty of tutorials around. In my experience, Vegas is the easiest video editor that can produce this level of result.

Keep in mind that project settings aren’t binding… you can change them any time you like.

And also in Options>preferences>Video and Preview Device …for example, should I be tuning the Monitor color profile? I just can’t find decent info on those variables. It might help those of us who are struggling to just get things done fast and clean, to use more of Vegas.

That’s for color management… using an ICC profile, if you haven’t set it globally, you can specify a profile for Vegas to use (if it’s globally set, you don’t need this). Select the “use color management” box, you can overide the OS defaults for color management. If you find an .icc profile for your Apple monitor, you can add it, as above, and select it here. Things will look better, and more specifically, they ought to look the same as they do with the Mac, assuming the color profile you find is close to what the monitor reports.

-Dave