Jump to content

Recommended Posts

I don't know quite where to post this question, so it might as well be here.

 

I recently came across the observation [ http://lowendmac.com/tech/g3g4.shtml , toward the end] that a 604e can handle 64 simultaneous processes, whereas the G3 can handle only 4.

 

This is a question for the technological literati: Are there any circumstances under which this superiority of the 604e over the G3 can actually show up? Or is single-threading the killer, and is this fact a good part of the sense behind the old claim that the G3 was 'optimized for the way MacOS works'?

Link to post
Share on other sites
Or is single-threading the killer, and is this fact a good part of the sense behind the old claim that the G3 was 'optimized for the way MacOS works'?

 

The MacOS no longer works the same as when it did when the G3 was introduced into the lineup, did anyone notice a premptive multitasking operating system appearing later?

Edited by Guest
Link to post
Share on other sites

I should add to my original, I suppose, that all this assumes that the piece on LEM is correct. However, I recognize the name of the author, who generally seems to know what he is on about. One also assumes that the LEM policy would be to exclude total crap from the archive....

 

I took a special interest in the following statements, illustrating the point about the ability of the 604e to handle many processes at once (from the LEM piece, though admittedly not specifically referring to the 604e and G3 question):

 

I've had the opportunity to work on a 603 and 604 of the same speed and on the same bus. . . here's the example of Photoshop and Netscape. When a filter is running on Photoshop - arbitrary rotation of a 109 MB color RGB photo scan, I cannot browse the internet with Netscape while Photoshop does the work in the background. With a 604, I can, and there is very little interference.

 

Similarly, the RC5 DES challenge is running on my PowerCenter 150 all the time, and the number of keys calculated rarely changes, even when I'm running several programs. I have never seen a delay in processing, since the OS handles multitasking and priority properly. In effect, the RC5 challenge seems completely transparent on my machine, while on another 603/180 you can definitely tell that the RC5 program is running. It's just a difference in how many simultaneous things the processor is designed to handle.

 

Nobody doubts that a 604e is slower than a G3, as this is something evident from a comparison of the two running at roughly the same clock speed (say, 200MHz versus 233MHz): all you have to do to confirm that is to stick an upgrade card in a PCI Mac and watch it boot. I also own an 8600/300 604ev, which I have run as a server for some years, and I know that this is a powerful chip (and one that is in significant respects as much like a G3 as it is like the earlier 604e), but it is no match for a 300MHz G3 with equivalent cache for running something like a photoshop filter. I have even installed X on such a machine via XPostFacto, and X actually runs tolerably well on it.

 

But what about those simultaneous processes in the 604e architecture, and does anybody know how it might be possible to test this properly and come up with some real numbers? (I am interested in numbers under classic MacOS, not OSX - and yes, I did notice its appearance, which came, however, well after the introduction of the G3.)

 

My reason for asking is that I have just stuck a dual 200MHz 604e in what was originally a single 604e 8600/200, and I would like to know a little more about the potential capabilities of the machine. I am not sure that the usual things one reads about these machines (always and in every respect inferior to a G3) are exactly right.

Edited by Guest
Link to post
Share on other sites

The G3 is almost always faster,that was the reason for the 9700 to use it.

 

The G3 Macs often come with a full megabyte of cache or some on chip cache,

like the 750cx.

 

The 604 machines can run Copland that supported preemptive multitasking only for background processes if it didn't crash(lol).

Link to post
Share on other sites

On one occasion I used for a whole afternoon a PowerMac 9600/350 running Mac OS X 10.2 (thanks to XPostFacto) and it felt snappier than the iMac G3/350 I had at the time (both machines had 256 MB RAM if I remember correctly). But admittedly it's just an empirical observation.

 

Cheers,

Rick

Link to post
Share on other sites

Here is what benchmarks I could find on the net:

 

http://lowendmac.com/ppc/g3card-mb4.shtml

 

Using MacBench 4 on my 8600/300 I get 728 for CPU and 775 for FPU under OS 8.1.

 

Looks like 950-1000 is the score 300 Mhz G3's get for CPU depending on the cache (512K or 1MB) no idea what FPU would be for 300 Mhz G3s.

 

 

Just for the heck of it I ran my Beige G3 MT (433 G3 1MB cache) on MB 4 and got 1443 for the CPU and 1081 for the FPU.

 

Anybody have a G3 300 512K/1MB cpu in a PCI mac to compare? I can email MB4 to anybody who needs it.

 

I would think FPU wise a 604EV at the same clock speed is comperable to a G3 at the same speed.

Link to post
Share on other sites
  • 1 month later...

Further to this, from MacKiDo, point #6:

 

"The floating point of the 603 and G3 had only a 32 bit multiplier (ALU) -- so for floating point multiplies it took two passes (2 cycles) to do a 64 bit floating point operations. The 604's had a full 64 bit multiplier, and so did better at floating point math performance than the others (up to twice as fast). The G4 has picked up the 64 bit ALU, so the G4 should have better floating point performance than the G3."

Link to post
Share on other sites

It wasn't really a competition or anything. Besides, both IBM and Motorola made 601s, 603s, 604s, and the PPC 750. IBM made all of the 601s and QFP-style 604s, whereas Motorola seemed to make a disproportionate number of the 603s, in both QFP and BGA packages.

 

Essentially, IBM said, "Hey, the 603's replacement kicks some ass and has a crazy TDP, so we're not going to bother with a proper successor to the 604." Which is fine, except that the 750 doesn't support SMP without extra hardware or software to make up for the chip's lack of cache coherency, so it wasn't exactly a server or high-end workstation chip. Plus there's the whole 32-bit ALU thing.

Motorola may have been participating in developing the 604's successor with IBM and had simply taken the torch by itself when IBM threw in the towel, or they may have decided that the 750's shortcomings were too much to bear and built it on their own. Whatever the case, the 7400 is essentially a 604 manufactured on a better process (and consequently, with higher clock frequencies), a new bus architecture (MPX), enhanced L1 caches, backside L2 cache support, and, most importantly, the introduction of the VMX subsystem.

Link to post
Share on other sites
  • 4 months later...

In my experience the 604e, and 604ev is considerably faster Floating Point wise than the G3. The G3 is after all nothing more than a refined 603e with better integer performance. My 9600/350 ran circles around a 300mhz G3 card in Strata for rendering. Norton System Info also showed the 604e kicked the G3's ass in Floating Point. To me, the 604e also felt more responsive than the 603e and G3 cpu's. The 604e was like a multicore CPU for it's time, it featured 3 separate ALU's on silicon not including it's complex FPU, load and store, or branch unit. Not until the G4 came around did I feel the responsiveness of the 604e again.

 

I loved the 604e, too bad the G3 followed the 603e design, and not the 604e :(

Link to post
Share on other sites

Ah the 604e CPU. How I loved it. It was designed to be a man's chip in a world of girlie CPU's.

 

Like all things that make me happy, it suffered an unfortunate and untimely death at the hands of youth and beauty.

 

God I loved the 604.

 

Good thing I have my 8600/604/300Mhz Tower.

Link to post
Share on other sites
  • 7 months later...

Howdy,

 

To summarize what's in pieces here, the G3 is an updates 603 core. The 604 is capable of dispatching more instructions per clock than the 603 or G3 and the FPU is definitely faster. The reason that the G3 is generally faster is that the G3 has an L2 cache controller built-in and therefore almost always has fast L2 cache (133 MHz to 200 MHz as opposed to 50 MHz for the 604(e) and 100 MHz for the 604ev) and almost always used a faster SDRAM bus (66 MHz SDRAM as opposed to 50 MHz (typical) FPM).

 

If you were to put a 604e onto the same speed memory bus with the same amount of L2 cache also running at the same speed as a G3, it'd be faster than the G3 at everything.

Link to post
Share on other sites

If it weren't for the fast L2 cache, yes. Depending on what you're doing the 1 meg of L2 on a 604ev card, even though it's slower, might turn out more helpful than 512k of L2 on a G3 card. Compiling, for instance, while not exactly the most common activity on PowerPC Macs, often benefits more from larger caches.

 

If your G3 card has 1 meg of L2, then it'd be faster in most things.

Link to post
Share on other sites
As fas as the G3 being optimized for the way MacOS works, thats sounds like marketing BS.

I asked one of the CPU guys at IBM about that, and he said, yes they did a lot of profiling of MacOS, and common opcodes were given hardware, while seldom used instructions were done in microcode. In other words, the G3 was a 603e tuned for MacOS. The 604 was a mainstream (for IBM) Power processor, with emphasis on FP, multiprocessing, and multitasking in a server environment. One was an engine for a Mustang, the other an engine for a Greyhound bus.

Link to post
Share on other sites
  • 6 months later...
As fas as the G3 being optimized for the way MacOS works, thats sounds like marketing BS.

I asked one of the CPU guys at IBM about that, and he said, yes they did a lot of profiling of MacOS, and common opcodes were given hardware, while seldom used instructions were done in microcode. In other words, the G3 was a 603e tuned for MacOS.

I read a Usenet post about a similar thing in comp.arch, but it was for the older Motorola 68k series.

 

It was the same thing, though-- you ran a profiler on your Mac which recorded opcodes, sent the data to Motorola, and they used it for further research and enhancement of future products.

Link to post
Share on other sites
To summarize what's in pieces here, the G3 is an updates 603 core. The 604 is capable of dispatching more instructions per clock than the 603 or G3 and the FPU is definitely faster. The reason that the G3 is generally faster is that the G3 has an L2 cache controller built-in and therefore almost always has fast L2 cache (133 MHz to 200 MHz as opposed to 50 MHz for the 604(e) and 100 MHz for the 604ev) and almost always used a faster SDRAM bus (66 MHz SDRAM as opposed to 50 MHz (typical) FPM).

 

If you were to put a 604e onto the same speed memory bus with the same amount of L2 cache also running at the same speed as a G3, it'd be faster than the G3 at everything.

Would you say the 604→G4 relationship mirrored the 603→G3 relationship?

 

One thing I've never really understood about microprocessors is why some are good for MP and others not. Was m68k a MP-capable design? Obviously 604s, G4s, and G5s were.

Link to post
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

Loading...
×
×
  • Create New...