More scanner-and-printer shenanigans

[Ben...]

Do Macs still use RISC processors? I read once (some years ago) that the slowest RISC processor is faster than the fastest IBM-style (Intel, etc) chip.


Thanks for the info everyone Where the hell was this information in all those billions of Googles I did?

They have Intel in them now. You can run Windows on any Mac with an Intel processor. Some older Macs can't because they have a different processor.

[Newc]

When did they switch?

[atomic charvel guy]

I used to wail a ton of chips and memory out into the world, and what I do remember
is that MACS always had a different style module for them as opposed to PC's, and it always had more margin in profit on the deals than on standard memory.
Then they joined the PC format eventually, and the laptops which probably take standard memory now as well were always proprietary back a ways. never could figure out why but i liked it as we could make decent bread off of mac memory deals.
i always did hear how they were better for graphics from day one . I can't support the claim cuz all i know how to do is post-lol

[Strangletooth]

When did they switch?

I believe the first was the G4, IIRC...

[Newc]

Oh, when they went to the PowerMac format?

I've always associated Macs with the RISC processor, though I admit the last thing I consciously read about them was back in 1983 or so. I do recall the "RISC vs IBM" statement, including the price difference.

However, if a RISC processor can stomp all over a Dual-Core 3GHz Intel chip.....

Then again there's the issue of "is anyone developing for the RISC architecture ?" - most of my programs are PC/Mac on one disc, but apparently they're "PowerPC/PowerMac/G4" oriented, so either way all I'm doing is swapping out OSs.

I could always go back to my NT4 system. I tried a thousand times and could not crash the system. Everything around it crashed, and when NT had enough it let me know, but I think I only got the BSOD twice in 3 years on it.

However, NT has drivesize and RAM limits that I disagree with.


I'll do the "1200DPI print vs 300DPI print" comparison Hellbat suggested in the morning when I get home.

[hippietim]

Basically a RISC chip is a very stupid CPU. It only knows how to do a few things. So to get it to do complex operations you must break the problem down into very primitive instructions.

RISC chips are fast at what they do if you are measuring pure single instruction execution time. However, it takes a LOT more RISC instructions to do what a CISC processor does in a single instruction. Add instruction pipelining to the mix and it turns out that a CISC chip will actually do useful things much faster.

The other problem with RISC CPUs is that since they require many more instructions they actually consume much more memory. One of the biggest performance bottlenecks in any computer is paging in and swapping out code. Also, when you blow the instruction cache on a RISC CPU you slow things down much more than on a CISC CPU.

[hippietim]

Newc - Mac's started on the Motorola 68000 series which were really great processors (and most definitely not RISC). My life would have been a lot simpler years ago if IBM had gone with the 68000 chip over the 8086 from Intel. The 8086 instruction set is retarded compared to a 68000.

Writing device drivers and kernel code with 8086 was so much more obnoxious than 68000 code. Fortunately these days nobody writes that shit in assembler anymore but back then you had to for performance.

[atomic charvel guy]

IBM stands for I Bought a Macintosh

[rjohnstone]

IBM stands for I Bought a Macintosh

Being a former IBM employee, we had a few of them.

I Blame Microsoft
I Blame Management
I've Been Moved
I've Been Managed
etc...

Pat, The PowerPC 601 chip was the first RISC chip to go into an Apple computer back in 1992.
Followed by the 603 and 604 series.
Later came the G3, G4 and G5.
Apple switched to Intel chips in 2005.

The PowerPC based chips can still be found in the GameCube, Wii, PS3 and the XBOX360.

They are also still used in high end servers.

[Strangletooth]

Being a former IBM employee, we had a few of them.

I Blame Microsoft
I Blame Management
I've Been Moved
I've Been Managed
etc...

Pat, The PowerPC 601 chip was the first RISC chip to go into an Apple computer back in 1992.
Followed by the 603 and 604 series.
Later came the G3, G4 and G5.
Apple switched to Intel chips in 2005.

The PowerPC based chips can still be found in the GameCube, Wii, PS3 and the XBOX360.

They are also still used in high end servers.

Thanx for that bro, I thought the G4 was the first to use an Intel, I thought anyway , I may have to crack her open and have a looksee inside. That's the only reason I was thinking that. I don't recall a mac having an Intel prior to the G4. I guess I could've looked it up, but I was just going off the cuff here.

[Spivonious]

As has been pointed out, the only difference between a Mac and a PC these days is the OS and a couple hundred bucks

Newc - as Tim has explained, a 32-bit OS (95, 98, NT4, ME, 2000, XP) can only address 4GB of memory total, since 2^32 = 4,294,967,296.

If you're going to be working with huge images (if Tim's calculations are correct, then 5GB images), you will definitely benefit from moving to a 64-bit OS and buying more RAM. You can get Vista Home Premium 64-bit OEM for $110 on www.newegg.com. Just make sure that you get a 64-bit version of Paint Shop Pro or it won't make a difference. I don't know if it exists. Adobe has a 64-bit version from CS4 and newer.

[rjohnstone]

The G4 is not an Intel chip.
It's a PowerPC chip.
Just like the G3 and the G5.

[dg]

The problem here is not that Newc doesn't have a supercomputer running a 64 bit OS. He's generating a file that's way bigger than he needs. Why? I have no idea, but one guess is that maybe he's trying to match his his image resolution to his printer's DPI setting, and they are two completely different and independent things. Printer DPI relates to ink droplet size. You can set that as fine as you want no matter what your image file's resolution is. Image resolution is often called DPI, but it should be called PPI so people don't get confused. Here are two links that explain this better than I can, but the bottom line is that an image resolution of 300 pixels per inch is as fine as you should need to send to your printer no matter what ink droplet size you have set it for.

http://www.ridgewoodcameraclub.org/d...Resolution.pdf

http://www.earthboundlight.com/photo...esolution.html

more:

http://www.google.com/search?hl=en&r...&start=10&sa=N

[Spivonious]

But DPI is always dots per inch regardless of whether the dot is a drop of printer ink or the dot is a pixel on the screen or the dot is a point in a bitmap. If he wants to print every pixel that he has scanned then he should match the two settings.

The real question though is will it be noticeable? Probably not. 1200 dpi is extremely fine detail. 300dpi should be fine unless you're getting very close to the posters he's printing. It's akin to the 1080i vs 1080p debate. One of them has twice as many pixels, but if you're sitting a certain distance away from the screen, the human eye physically cannot pick up on the difference.

[hippietim]

It is true that scanning at 1200 DPI is pointless if you can only print at 300 DPI. However, if you are planning on enlarging the image then higher resolution is a good thing.

By increasing the printed size you are really just decreasing the DPI. Your scanning DPI should account for the scaling and output DPI:

Scanning DPI = Printer DPI x Scale Factor

So if you scan an 18" x 36" image at 1200 DPI and decide to print it out at 24" x 48" then you are really just changing the effective DPI to 900. If your printer can only do 600 DPI then scanning at 800 DPI should get you the same results. If you print at 300 DPI, scan at 400.