Not to jump into the fray, but coming from someone who has been looking for a job via their sig on/. for at least 2 months, it sort of loses some effect.:)
Two years ago, I'd have agreed. In this economy, no.:-)
I second this. I found my old copy of Independance War a few months back, and was all excited about playing it, till I remembered it was Glide or Software. I was THIS close to digging out my Banshee card....but it just didn't feel right plugging it in beside my Geforce 3....
Yes, but film/video/whatever are also exposure over time, where as a computer frame (as in FPS) is a snapshot in time.
In other words, on film or video, if an object traves from one end of the screen to another in the space of two frames, you'll see a blur on the film. For something like quake, you'll see the object on one side of the screen, then the other side. More FPS for the same action will allow for a finer granularity; if the two frames show the object on the sides of the screen, left and then right, then five frames per second will show the object at 0, 33, 50, 66 and 100. Ten FPS will show the object at 0, 10, 20, 30, 40, 50, 60, 70, 80, 90 and 100; both in the same space of time. This winds up giving you that visual continutity that film gives you automagically.
FYI, I stopped using Linux in a production environment when Mandrake 7 and 8, and RedHat 7.2, when told to run a cron job at 4 AM every morning, using Vixie crond, would run it at perfectly random times, day or night, and multiple times, day or night, on several different machines.
Linux desperately wants to be in the same world as Windows NT and Solaris, but people like you then complain long and loud when one tries to apply the same rules.
Ahhh, if only microsoft let the general public have access to weekly kernel builds and the sourcecode to come along with it..
Bullshit. With OSS, it's "Release early, release often." With Microsoft, it's "They obviously didn't even bother to test this. Typical Microsoft slipshod work. Haven't they even heard of QA?
I can't help but notice you're using a pseudonym. You wouldn't be doing this unless you have something to hide; why else would you not give your real name?
People downloading kernels from kernel.org, particularly in the first few days of a release, are part of the QA process, not the ultimate beneficiaries of one.
The funny thing is that if you substitute 'buying Windows' for 'downloading kernels' then you're saying the same thing about microsoft, only instead of being earnestly serious, you're being snide and condescending.
A fix was released in merely hours. I don't see Microsoft doing that.
Well, I do seem to recall/. yelling at microsoft for Code Red, which went wild WEEKS after the Microsoft patch which fixed it was out, and MONTHS after the IIS HowTo which told you how to avoid the whole mess in the first place was out.
Why would you want to? Sid Meier has absolutely nothing to do with what Infograms does or doesn't do. The best Sid can do is say 'wow, that sucks. Remind me NOT to renew our publishing contract when it comes up.'
No, actually, people lost data. Oh, and my post wasn't a troll, it was a shout of 'the emperor has no clothes!' which will, of course, lead to me getting lynched by the imperial court.
If only this was Open Source Software, the source code could have been examined by thousands of highly motivated and intelligent hackers, who would have noticed the problem immediately.
Wait....
If is such an advantage, why does it not appear in PCs that cost more than $700?
It's the HOW of it.
Going back to my example. You've got a pail full of water, an empty pail, and a garden hose to transfer the water from one bucket to the other. That's your standard PC.
You've got one full pail, and a garden hose to drink through. That's your 700 dollar PC.
You've got one full pail, and a firehose to drink through. That's UMA.
In other words, there's a difference between sharing RAM, and having RAM designed to be shared at maximum efficiency.
Yup. Our ACD system (call centre queues and the like, ACD standing for Automated Call Distribution) was a single board Pentium 133 w/32 megs RAM running OS/2. You booted it up, watched it load a bunch of DOS drivers, boot to the OS/2 desktop, then autorun the phone app.
PC = lets say 64 megs of RAM, 64 megs of Video RAM and an AGP bus that lets you copy between the two (think having two buckets and a hose)
Unified Memory Arch = 64 megs of RAM which both your video chips and processor are plugged into.
Note that originally AGP was designed to work as a poor man's UMA; the card would go directly to system RAM over the AGP bus, obviating the need for on-board RAM. Nobody actually does this.:-)
So lets say you want to create a texture on the fly. On the PC, your processor creates it, dumps it in RAM, and it's then copied (and this is the bit that takes time) over to the video card's RAM, where it's read and displayed.
On a UMA machine, the processor creates it, dumps it in RAM, and the video card reads it and displays it.
Now, IIRC, the PlayStation2 does this the exact opposite way; lots of little bitty cup sized RAM caches, and a firehose of a bus. The idea being grab some data, use it, get rid of it. You need it later? Just grab it again, you've got bandwidth to spare.
I would far from quantify it from 'wasted.' It's a tradeoff. And the next level up, stripes across mirrored sets, you use 2 drives to store one drive worth of space.
Basically, though, if you need to ask, you probably don't need RAID.:-)
Oh hell yeah. I still remember my sweet PERC2 cards on my dell servers (and they're far from the best things on the planet) with 128 megs on the card itself, several days worth of battery backup for said RAM, and all sorts of co-processing goodenss.
Slashdot Mirror
So throw in lossless compression; I believe it averages, for music, 1:4, as opposed to "decent" MP3 which averages 1:12.
Unless it was Apple. Then it still would destroy hard drives. :-)
I second this. I found my old copy of Independance War a few months back, and was all excited about playing it, till I remembered it was Glide or Software. I was THIS close to digging out my Banshee card....but it just didn't feel right plugging it in beside my Geforce 3....
Yes, but film/video/whatever are also exposure over time, where as a computer frame (as in FPS) is a snapshot in time. In other words, on film or video, if an object traves from one end of the screen to another in the space of two frames, you'll see a blur on the film. For something like quake, you'll see the object on one side of the screen, then the other side. More FPS for the same action will allow for a finer granularity; if the two frames show the object on the sides of the screen, left and then right, then five frames per second will show the object at 0, 33, 50, 66 and 100. Ten FPS will show the object at 0, 10, 20, 30, 40, 50, 60, 70, 80, 90 and 100; both in the same space of time. This winds up giving you that visual continutity that film gives you automagically.
This was a year back, and I went through the appropriate support forums, such as they were. It wasn't worth my time to deal with, though.
FYI, I stopped using Linux in a production environment when Mandrake 7 and 8, and RedHat 7.2, when told to run a cron job at 4 AM every morning, using Vixie crond, would run it at perfectly random times, day or night, and multiple times, day or night, on several different machines. Linux desperately wants to be in the same world as Windows NT and Solaris, but people like you then complain long and loud when one tries to apply the same rules.
I can't help but notice you're using a pseudonym. You wouldn't be doing this unless you have something to hide; why else would you not give your real name?
Why would you want to? Sid Meier has absolutely nothing to do with what Infograms does or doesn't do. The best Sid can do is say 'wow, that sucks. Remind me NOT to renew our publishing contract when it comes up.'
No, actually, people lost data. Oh, and my post wasn't a troll, it was a shout of 'the emperor has no clothes!' which will, of course, lead to me getting lynched by the imperial court.
If only this was Open Source Software, the source code could have been examined by thousands of highly motivated and intelligent hackers, who would have noticed the problem immediately. Wait....
So the 1996 implementation of tech A is beaten by the 2000 implementation of tech B. What about the 2001 implementation of tech A?
Yup. Our ACD system (call centre queues and the like, ACD standing for Automated Call Distribution) was a single board Pentium 133 w/32 megs RAM running OS/2. You booted it up, watched it load a bunch of DOS drivers, boot to the OS/2 desktop, then autorun the phone app.
PC = lets say 64 megs of RAM, 64 megs of Video RAM and an AGP bus that lets you copy between the two (think having two buckets and a hose) Unified Memory Arch = 64 megs of RAM which both your video chips and processor are plugged into. Note that originally AGP was designed to work as a poor man's UMA; the card would go directly to system RAM over the AGP bus, obviating the need for on-board RAM. Nobody actually does this. :-)
So lets say you want to create a texture on the fly. On the PC, your processor creates it, dumps it in RAM, and it's then copied (and this is the bit that takes time) over to the video card's RAM, where it's read and displayed.
On a UMA machine, the processor creates it, dumps it in RAM, and the video card reads it and displays it.
Now, IIRC, the PlayStation2 does this the exact opposite way; lots of little bitty cup sized RAM caches, and a firehose of a bus. The idea being grab some data, use it, get rid of it. You need it later? Just grab it again, you've got bandwidth to spare.
Yeah, lord knows, for example, that you can run out and buy an Athlon with a Unified Memory Architecture.
I would far from quantify it from 'wasted.' It's a tradeoff. And the next level up, stripes across mirrored sets, you use 2 drives to store one drive worth of space. Basically, though, if you need to ask, you probably don't need RAID. :-)
Oh hell yeah. I still remember my sweet PERC2 cards on my dell servers (and they're far from the best things on the planet) with 128 megs on the card itself, several days worth of battery backup for said RAM, and all sorts of co-processing goodenss.
Well, as you say, if you're outputting to a TV, your limited to less than 640x480 for resolution.