Difference '030@50MHZ and '040@50MHZ

[stefcep2]

I have had all of these cpu's at some time.

Speed often comes at the cost of compatibility: an 030 will run pretty much anything you want at good speed with an fpu, the 040 will run less stuff, and the 060 less still.  TBH the sort of stuff 040 and 060 was made for you can do better and faster on cheap PC running winuae, unless you HAVE to have an Amiga to do it, its your money.

Getting to bench marks: if you run software optimized for it the 060 can be 4 times faster than 40 mhz 040, which is 4-8 times faster than 50 mhz 030.

If you do CPU intensive stuff that uses floating point arithmetic, then you need to have the correct libraries, the correct cpu command and a patch: cyberpatcher for 68004/60 made by phase5, 680040/60.library for GVP and oxypatcher for everyone else.  There's also hsmathlibs which speed things up even more for stuff that the other patches don't patch.  These patches make a HUGE difference to rendering and image processing.  Some phase5 boards actuall speed up chip ram speed as well.

Memory speed does vary and can make a difference:  I benchmarked my memory speed of an Apollo 1240@40 mhz in AIBB and it was faster then a published memory speed benchmark for a PPC/68040 board.  And i noticed it too: booting was very fast as was copying to ram: disk, faster than my Cyberstorm 2 68060 on my A4000.

My advice: stick with an 030.

[stefcep2]

cantido wrote:
Raytracing etc isn't going to benchmark the CPU. Instead you'll be benchmarking CPU + Amiga hardware + OS + Misc software. To benchmark the CPU opposed to "the system" you'd need to write a benchmark that takes over the CPU and runs from memory that doesn't need any waitstates.

True and this can be done and is often quoted, and you get pretty meaningless results. IMHO its "benchmarking CPU + Amiga hardware + OS + Misc software" that matters to the end user

[stefcep2]

cantido wrote:

Crumb wrote:
@bloodline

MIPS = Meaningless Information Processor Speed.

I don't see why it's "meaning less". It shows how many instructions can be executed per second. If those instructions are of equal value the processor with the highest value will be "faster". It'd be silly to compare two totally different processors of course.

At what point do they become "two totally different processors"?  Is an 030 "totally different" to 68060?  Well other than running nearly the same 68k code as opposed to x86 code or alpha code, they share nothing else in common in terms of transistor number, pipelines, cache, fpu instruction set, design, number of executed instructions per cycle. Turn the caches off on an 040 or 060 and they execute the same instructions slower than 50 mhz 030.  Turn the caches on and they may not execute at all!! AIBB is good (but not for 060) because it does things like rendering a beachball and timing it and yeah mips gives very misleading results.

[stefcep2]

Crumb wrote:
@Kin-Hell

overclock your MKII 060 to something higher than 56Mhz (e.g. 60Mhz) and you will notice it a lot faster (I've heard it's due to the fact that the turbocard will synchronize better the motherboard)

Yea and you then will get unreliable scsi if you have it.

[stefcep2]

Crumb wrote:
@stefcep2

It shouldn't cause problems at 60Mhz.

i read somewhere that anything above 50 mhz is a problem.  Its years though, but i remember thinking about doing it, and was advised against it for this reason

[stefcep2]

shoggoth wrote:

stefcep2 wrote:
At what point do they become "two totally different processors"?  Is an 030 "totally different" to 68060?  Well other than running nearly the same 68k code as opposed to x86 code or alpha code, they share nothing else in common in terms of transistor number, pipelines, cache, fpu instruction set, design, number of executed instructions per cycle. Turn the caches off on an 040 or 060 and they execute the same instructions slower than 50 mhz 030.  Turn the caches on and they may not execute at all!! AIBB is good (but not for 060) because it does things like rendering a beachball and timing it and yeah mips gives very misleading results.

I think we need to make a distinction between CPU(+memory) benchmarks and system benchmarks. The former is of interest to those using CPU intensive stuff like rendering, while the latter takes system bottle necks into the equation.

Personally, the mips figure makes sense to me since I get a rough idea of what kind of performance my code can get when only accessing RAM.

Not a coder but i get your meaning.

[stefcep2]

That might be true for the DCE A1200 boards.  My experienmce was with the cyberstorm 2 for A4000 with cyberscsi module.  i am pretty sure that with the cyberstorm boards overclocking caused problems with the the scsi board.