Earlier I posted an e-mail, inquiring into Pierre-Olivier Latour's noticeable absence of contributions to this site. His last post was 2/11/08. I was hoping for some professional, polite comment from [EMAIL PROTECTED], if for nothing else than to wish him a 'Bon Voyage." I've been involved in research into Emergent Contextual Relations/Relational-Hierarchical Semantics and their deployment on a neighborhood of 9 screens (think tic-tac-toe). I briefly mentioned this to M. Latour (see below) and, since the work is largely unpatentable, I had no problem with sharing this. However, In addition to wishing Pierre 'good travels,' I am concerned about a device (patentable) which I explained to him via private e-mail. I felt safe that I could trust Apple to not adopt this, or any related devices, but now I grow concerned, particularly if M. Latour is no longer an Apple employee. It was a mockup made with black pipe-cleaners and tooth-picks (these were painted white--strictly aesthetic) in which a single keystroke could be propogated mechanically to 9 keyboards--concurrently and simultaneuosly in parallel. This was only after I reviewed the prices of existing 9 keyboard KVM switches and that horrible limitation known as BlueTooth Pairing (I am hoping that there will one day be a "BlueTooth Neighboring", but time is of the essence. So, if you see such a device manufactured in the near future--I thought of it first! Please find my earlier e-mail below:
Bonjour Monsieur Latour, I'm sorry, but, in truth, I don't speak French very well. I did attend the Tiger "rollout" in Chicago a few years ago and had a wonderful chat with Guillermo Ortiz, who gave the keynote, and I believe you presented various graphics related topics. I did want to briefly express that I've had great pleasure in reading this mailing list over the past few years and always appreciate your timely and well-informed, professional communications. Quartz Composer under Leopard--mon dieu! It sure has grown more robust. Generally, I've been doing independent, self-funded research on developing a neighbor-centric workstation. Practically, this means 9 19" display screens arranged in a collaborative matrix. Tagline--A Bigger Mousepad or a Better Mousetrap? Is there a market for this? I can only say that I spent some time with a Next workstation years ago and I thought it was best-in-class. For this, steve jobs' company went bankrupt and now for designing the thinnest client imaginable (the iPod), Apple has $15 billion in the bank. Truly, this is a nice device, I have 2. And more importantly, it's hard to steal hardware. (a truly sad observation on human nature). At any rate, Next made it's way to Mac OS X and I believe that the best is yet to come. I think that I've only posted 2 or 3 e-mails to your list and earlier this morning I found myself composing a slightly sarcastic e-mail. I found it humorous, but thought that it might not be appreciated by other contributors who sound like they have real problems to discuss. I try to stay out of vague, performance related issues. They all seem to boil down to a graphics pipeline bottle-neck, which a PowerMac would improve, or 9 collaborative graphics pipelines. Personally, I would say that neighbor-centric processing represents a significant paradigm-shift, but "Wired" magazine informs me that "paradigm-shift" is a "tired" expression. I'll have to re-read Thomas Kuhn. I promised earlier to be brief and notice I've been babbling, so I'll conclude, with the potentially offending e-mail following. A tout.... APPLE ROCKS! Sincerely, tom emerson daronatsy "sometimes it seems like it's rainin' all over the world":-( ------------------------------------------------------------------------------------------------------------------------------------------------------ I've got a mac mini (gma950) and I notice that I can pan a lot faster if I just make my "viewport" smaller (and probably faster yet if I use grayscale). Both of these don't tax my graphics pipeline (aka-transfers to vram) as heavily as 32-bit full-screen. I've got the mini hooked up to a 60" Aquos, so a 100x100 viewport is still pretty big (I haven't calculated the dot pitch), but the pixel resolution is the same as the 24" Cinema display (1920x1200). I'm ecstatic over the performance of my mini--this was under $500. Realistically a PowerMac might be better, or a neighborhood of minis. Moore's Law II (with apologies to Moore)--never spend more for your monitor than you do for your computer:-) On Tuesday, August 12, 2008, at 09:34AM, "Carl Rohumaa" <[EMAIL PROTECTED]> wrote: >Thanks, >Yes I get more FPS in asynchronous mode in fact over 30 BUT it still >is not smooth like in QT. Still has a bit of a judder. I can even see >the QT version running smoothly full size with QC version juddering in >the background! Would have really liked this to work for a video >installation I am creating for a show in a couple of weeks. I can even >see the QT version running smoothly full size with QC juddering in the >background! > > >On 12 Aug 2008, at 14:47, Christopher Wright wrote: > >>> I have a video (1280 x 720 Photo JPEG) that plays back fine in >>> Quicktime but only achieves 18-20 FPS in Quartz Composer so I get a >>> jerky motion. I would really like to play this back in QC. My >>> Laptop is a Powerbook Pro 2.16GHz 2GB the GPU is RadeonX1600. Just >>> interested as to why this might be. >> >> The Movie Loader patch has terrible performance :( (We are fighting >> with this too on a project) -- for slightly better framerates, try >> opening the inspector panel, and making the patch run in >> "Asynchronous" mode. You'll lose the ability to jump to arbitrary >> times, but you'll gain the ability to play audio from the movies -- >> depending on what you're trying to do, this may be completely >> useless and unhelpful, or it might get you a few more fps. >> >> <mode type="nerd"> >> The movie loader patch performs poorly perhaps in part because it >> requires the composition's frame time each frame. Because the >> composition often runs at a different framerate than the movie, the >> movie loader has to wait till the next composition frame to see >> exactly when it is. Because of this, it can't simply decode "the >> next frame", because that's not correct with respect to time. >> Quicktime has an advantage here, in that it can run at whatever >> framerate it wants, and can just decode the movie sequentially >> (jumping to arbitrary points is possible, but not the common case), >> which is pretty cheap for most codecs. seeking to a given frame/ >> time though is maybe a bit more expensive, and QC effectively does >> this every frame. >> </mode> >> >> -- >> [ christopher wright ] >> [EMAIL PROTECTED] >> http://kineme.net/ >> > > _______________________________________________ >Do not post admin requests to the list. They will be ignored. >Quartzcomposer-dev mailing list ([email protected]) >Help/Unsubscribe/Update your Subscription: >http://lists.apple.com/mailman/options/quartzcomposer-dev/aram1003%40mac.com > >This email sent to [EMAIL PROTECTED] > > _______________________________________________ Do not post admin requests to the list. They will be ignored. Quartzcomposer-dev mailing list ([email protected]) Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/quartzcomposer-dev/archive%40mail-archive.com This email sent to [EMAIL PROTECTED]
