Half Life 2 CPU Scaling Tests are Done

When November came around, I was a bit shocked by how many product launches we had happening that month - so close to Christmas. I shrugged it off as a fluke and we all worked hard to get the reviews out and everything done on time; December was supposed to be a much more relaxed month.

Then ATI gave us a call and said that they were launching 5 new products - ok, no big deal. Then NVIDIA sent word that they were finally launching PureVideo...on the 20th of December. I figured that was it, but no, I've actually got more hardware being delivered to my parent's place in NC so I can benchmark over the holidays. I told myself I wouldn't leave CT for the holidays until I was done with my work...well it looks like my work will be leaving with me :)

I'm finally done with the Half Life 2 CPU scaling tests now that I'm pretty much done with all interruptions. I've got to write it up but that'll have to happen while I'm down in NC. Now last time I said I was going to do work while spending time with the family in NC, nothing got done. I'm not sure what will be different this time, but something will...I swear. Honestly I'll feel a million times better once I can get this final part to the Half Life 2 CPU investigation done. I was quite tempted to turn it into a massive "impacts of CPU performance on next-generation game engines" article but a basic desire to get it done and out the door prevented that from happening. It's still a bit beefier than I originally expected it to be, but not as overwhelming as once thought. There are still some interesting points to make and a lot to take from the numbers I've run, so I hope you all will enjoy it.

Gigabyte is sending me their 6600GT single-card SLI board while I'm down in NC, so if all goes well I'll be able to do a quick preview of it from warmer climates. I will be bringing some of my testbed equipment with me for said purposes.

I've got an official date for a Mac article - 1/5/05. That's all I can say for now.

Because of my highly unpredictable work schedule we decided to drive down to NC for the holidays. Maybe it's just that I haven't subjected myself to the crazy drive from NC-to-CT in a while, but I swear I must've told myself I'd never make that drive again somewhere in the past several months.

Sorry if I'm seeming a bit more random than usual in this blog, I ended up coming down with some sort of a cold-thing earlier this week. Me getting sick in CT was a long time coming, staying up for around 40 hours straight to finish that PureVideo article sealed the fate on my short-term health. The PureVideo article didn't turn out as well as I would've liked it to, but I think it conveyed the strengths of PureVideo as a de-interlacer, but nothing more.

The fact of the matter is that with a single registry edit ATI is able to offer similar CPU utilization to NVIDIA under WMV9. What NVIDIA offers is better de-interlacing quality but I'm still not totally convinced that NVIDIA needed ~20M transistors to achieve that. I think what's also disappointing is that we've waited this long to get WMV9 decode acceleration and even with it supported through the driver, we *still* have to wait on Microsoft for an update to Windows Media Player 10.

We finally had our first "real" snow in CT the other day. The cars have appropriate shoes so driving wasn't a problem, but I'd forgotten how dirty everything gets when it snows. I probably used the windshield washer fluid more in the past 2 days than I have in the past year combined. Luckily most of the slush and other randomness appears to have vanished from the ground...the day before we leave :)

I've gotta get back to dumping things on my laptop, but I may make another post before we leave late tonight.