Busy Week

Not sure how much posting will get done this week as Silicon Valley heads into the final stretch before the Christmas break. We still have a lot of business to close, and less & less time now in which to do it.

Therefore, "here's a test card and some music" as they used to say on the BBC when there was no programming on (remember those days?) in order to keep you occupied in the meantime.

OK, then ... pelicans flying over some flamingos, all delivered in absolute silence. Happy now?

B&W Technique for High Key

There's a great web site from Diane Varner that has some very interesting examples of toned shots she has done in CS3. Try as I might, I can never quite get this to work well for me, but will nevertheless keep struggling. I like to blame this on being colour-blind ... but that's not really quite fair! Being vision-challenged, yes, but not in that way.

Anyway, I was reading through her post there on post-processing technique and saw that she'd used a different B&W conversion approach that I hadn't seen before: the Gradient Map.

The advantage of this approach is that you can control the mapping between brightness levels in the original and the resulting grey-scale image. Furthermore, you can do very interesting things like such as black/white/black conversions, with the highest levels going back to black instead of blowing out.

The original (this one is close enough) was useful because it had a fairly constant blue background that could be reliably mapped into white in order to reset the feel of the shot, also allowing some leeway in transforming the real highlights to help add definition and delivering more of a high-key feel.

Detailed instructions? Alas, I can't give the full recipe for this one because I did it late at night and forgot to save the full layer set before flattening - doh! I think I first boosted contrast with USM at a high-radius/low percentage level before doing the gradient map. When converting, I used the "metal" preset because that had a good starting point for the black-white-black approach. And curves, I'm sure there was a curve in there to do some global contrast adjustment. Somewhere.

Anyway, give it a go, and if I can dial-things in a bit more accurately I'll post on this topic again. Please though let me know if you can craft a more reliable recipe and I'll repost it to everyone.

To be or not to be ...

Start-ups can be tricky. It all comes down to whether or not you can find or form a viable market segment that can support the growth of your new technology or product to reach profitability (and then onwards and upwards from there of course!) And remember, this is only *after* you have spent many millions of investors' dollars just to get that first product version out the door. Look across Silicon Valley,though, and the overall odds aren't great for any sort of real success. Roughly a third of new start-up companies will fail outright, another third will reach some sort of neutral (+ or - a few percentage points) outcome and the the final third will generate some level of return for investors and hopefully the company employees. In this day & age there are very few "knock it out of the ball park" outcomes that can help offset the failures and generate big returns for VC funds. That tends to rack-up the heat on portfolio companies to produce, and preferably produce now.

No wonder it's hard to map out any kind of well delineated future path capable of supporting a solid operating plan that a start-up can work to. I know many start-up CEOs and, to some degree or another, we all face exactly the same problems: timing and suitability. Can we find/build/bully a market into existence before the VCs tire of the investment required and the funds dry out, and can the business support a long term, growing revenue stream big enough to open up the opportunity for a decent exit? (Of course, we're not talking about the YouTubes of this world but rather the wide range of other start-ups creating new tools, business models, etc. that comprise much of the "real-world" environment we see in SV. Nothing against these "eyeballs only" operations, but that's not where most of us earn a crust.)