Cringely on Cells

February 15, 2005 03:00 by keithkaragan
After reading Cringely's ' What Carly Will Be Missing ' article yesterday my hope for the IT industry is even further revitalized. First the Firefox browser, then the mac -mini, now a true leap in the techo-industrial-revolution. What's truely amazing about the cell (nice overview here ) is that it may really turn the industry upside-down, spurred by the gaming industry - Sony! I've long held the opinion (yeah, I know your saying 'sure you did') that the biggest threat to Microsoft was Sony, die to the massive amount of money they invest in research and their mindset as a consumer product company - they understand consumers (although some may say otherwise due to some choices they made with the mini-disc), they get the 'it just has to work' attitude they ('we') have. Press a power button and it on, press again and it's off - no drama, no waiting. Controls need to be obvious and immediate. THey've toyed in the PC industry for awhile now - and with nice results, I love my VAIO laptop - it really is one of the best laptops I've ever owned - and it was not very expensive and has a 16" screen. Playstation may hold the power to dethrone the current major software kings in a very short time. Apple may find Sony to be their knight in shining armor as well.
In the Inrto to Cell Processors, Blachford points out the current connection between the PPC proc as a controller for the vector processing cells. Not that other types of processors couldn't be used - but this is the current flavor and works (PS3 will apparently use this configuration). If Apple really wants the market (Read Cringely's previous article on the mac -mini's really purpose...) it laying at their feet. Their OS is mega-slick, they own the creative market (ProTools, Avid, Logic, Photoshop, etc) and it will get so much more superior in such a short time (regardless of their actions) once the makers of these apps adapt to utilize add-on cell processors - since these cells are optimized for the kind of processing these utilities do. Once integrated, with a lower price point they could dominate the market - especially if they license the OS to white box makers, keeping the high-end for themselves.
Microsoft is not out of the game - remember the PPC versions of NT and consider that XBox is PPC too - I wonder how much talent they have on staff that are versed in these technologies, and how well these folks are working with the .Net and Tools researchers ... as Blachford points out - this will be key since you need to have usable software to take advantage of the hardware. He also mentions that XML has not shown itself to be a great candidate for use on Vector processing systems ... .Net is all about XML, it runs throughout it, can this be the Achilles heel in MS's software platform as far as this is concerned? Or is there an easy solution to abstracting the XML aspects of .Net technologies to get more performance from .Net apps running on a vector system - .Net is platform independent (although most apps are utilizing aspects of the Win32 API, or some legacy COM through PIA libraries and Interop - this would need to change to some extent, I imagine).
I don't have a decent handle on the Apple dev tools (not Java, but Objective C and XTools) - hopefully I will soon enough, but I wonder how well this performs in the vector universe. I would thik it does well, given the palette of current apps (like Logic Audio) that seem to use parallel computing in real apps that real people use today. With Logic, you can apparently add more CPUs to the mix (other machines) and they'll work together in your recording studio. Another anecdote is the mac cluster - apparently it's very plug and play .... just like the cell processors - Hmmm, interesting? Add to that the Airport as as a ubiquitous wireless networking infrastructure for all things Apple and the support they've had for Gigabit ethernet and there is definitely a compelling argument that they have the right mindset and have acted upon it.
Windows is moving in some directions that seem to indicate that they are not ignorant here either though - Longhorn technologies are seemingly a step in the right direction, and their DRM focus would be seemingly strengthened here, and don't they have deals with Sony? Flextronic has been producing PPC based devices for MS for sometime now (and a new phone in the works, reports say), but where is the brain trust on PPC and MPUs outside of the XBox team(s)? Maybe it's there, maybe it's not - time will tell.
But if nothing else this evolution will shake the IT tree, with industry reaping the benefits in leaps in investment (Cringely gets into the logic of this with his talk of Venture Caps in his article) and we're off to the races, not even a war or political dissent is likely to stop this from occurring - The fact that this money needs to be invested will bring it on full throttle. What will be the apps and systems that run this surge? Mobile (likely), Media (definitely), but what else? Will we finally see leaps in the statistical / computational / modeled data analysis that we have tools for but not the breadth of implementation of?
Also, on a personal note, what does this mean for the IT consultants and programmers? For sure, we'll be climbing the learning curve again - re-tooling for the next ride on the roller coaster, figuring out what we can and can't do with all this cool stuff. I can't wait!
Digg It!DZone It!StumbleUponTechnoratiRedditDel.icio.usNewsVineFurlBlinkList

Be the first to rate this post

  • Currently 0/5 Stars.
  • 1
  • 2
  • 3
  • 4
  • 5

Related posts

Comments are closed