Overview Features Coding ApolloOS Performance Forum Downloads Products Order Contact

Welcome to the Apollo Forum

This forum is for people interested in the APOLLO CPU.
Please read the forum usage manual.
Please visit our Apollo-Discord Server for support.



All TopicsNewsPerformanceGamesDemosApolloVampireAROSWorkbenchATARIReleases
Performance and Benchmark Results!

Is Vampire Faster Than Classic PPC Cards?page  1 2 3 4 

Chris Sanz

Posts 25
01 Feb 2018 04:20


Forget multi users. It's a PC so I would still like a password login mechanism to keep out prying hands.


Vojin Vidanovic

Posts 770
01 Feb 2018 06:52


Chris Sanz wrote:

Forget multi users. It's a PC so I would still like a password login mechanism to keep out prying hands.

There were couple tools since 1995, from "keep my sis out"
EXTERNAL LINK 
To tries of multiuser implementation
EXTERNAL LINK


Gunnar von Boehn
(Apollo Team Member)
Posts 6207
01 Feb 2018 15:38


Vojin Vidanovic wrote:

By design 603/604e were early PPC experiments,
and while gained ground, havent been any real revolution.

Do you refer to the Accelerators using them, or to the CPU chips?

Looking at the CPUs I would regard the PPC 603 as solid CPU, and the 604e as strong PowerPC.
The 604e is in several areas architectural stronger than the G3 PowerPC, and the 4x0 chips used in SAMs, and also in areas stronger than the PA-Semi chips used in latest machines.




Vojin Vidanovic

Posts 770
01 Feb 2018 16:18


Gunnar von Boehn wrote:

  Do you refer to the Accelerators using them, or to the CPU chips?

Its more impression, or lack of it, by early PowerPC Macs that have used it. Was it a "Pentium killer" (as its x86 class comperitor) I really dont know.

In Amiga land, it didnt "quite live up to its potential" (that would be impression) but cards were impressive for its time offering SCSI, fast RAM slots and gfx card option in some versions. They were in fact last breath of Amiga market, despite high price.


Thierry Atheist

Posts 644
01 Feb 2018 16:24


The Amiga PPC accelerators probably didn't have RAM that was fast enough for the CPU using it.


Gunnar von Boehn
(Apollo Team Member)
Posts 6207
01 Feb 2018 16:52


Thierry Atheist wrote:

The Amiga PPC accelerators probably didn't have RAM that was fast enough for the CPU using it.

The POWERPC chips are good and the PowerUp cards also, and the RAM they use is also good for its time.

APOLLO is just faster, thats all.


Will 'Akiko' G.

Posts 9
02 Feb 2018 08:41


Gunnar von Boehn wrote:

  The 604e is in several areas architectural stronger than the G3 PowerPC, and the 4x0 chips used in SAMs, and also in areas stronger than the PA-Semi chips used in latest machines. 
 

 
  Very true, the 604e and 604r later resulted in the 620 and the POWER chips, which are quite strong designs. The MPC750 (G3) lacks one integer unit and can execute less instruction in parallel in comparison to the 604. But it has higher clocks and a much stronger FPU. The 4x0 SoCs and the PA-Semi are embedded designs, so basically these eX00 cores, which are not very powerful but also don't draw much power.
 
  So, if you wonder why I keep coming up with this stuff. I was part of the team who ported KVM (Linux kernel based virtual machine) to PowerPC/POWER about 10 years ago. For this I literally had to read the whole reference manuals of 603, 604, 750, 7400, 970, e300, POWER4 and POWER5. I still have most of the hardware.
 
 
Vojin Vidanovic wrote:

  Its more impression, or lack of it, by early PowerPC Macs that have used it. Was it a "Pentium killer" (as its x86 class comperitor) I really dont know.
 

 
  They were, but mainly because they had higher clocks. I remember some ATX bases MPC604 boards, where you could run WinNT 4.0. But during this time it was more fun to use Alphas or better HP-PA (which are stronger than Alphas).
 
 
Gunnar von Boehn wrote:

  The POWERPC chips are good and the PowerUp cards also, and the RAM they use is also good for its time.
 

 
  I wouldn't say that about the powerUP cards. Putting two architectural totaly different cpus on a shared memory bus is a bit insane. To keep them performing well every cpu has to get its own memory, both connected by a "mailbox system" (small shared memory with some synchronization mechanism). Just like it is done today in SoCs who have an embedded DSP or FPGA next to cpu part. In the case of the powerUP cards I wonder if they tried to sync data in the caches of both cpus.


Gunnar von Boehn
(Apollo Team Member)
Posts 6207
02 Feb 2018 09:08


Will 'Akiko' G. wrote:

Very true, the 604e and 604r later resulted in the 620 and the POWER chips, which are quite strong designs.

If we look at CPU relationship then one can draw this overview:

68000 and 68010 are nearly the same chip
The 68010 is based on the same design as the 68000, but has small improvements over the 000.

The 68020 is a complete different CPU than the 68000.
But the 68030 is nearly the same chip as the 68020.

The 68040 is a complete new architecture and is internally very different to the 68020/68030.

The 68060 is again a complete new architecture and is internally totally different to the 68040.

The 68080 brings many advancements, as 64Bit, AMMX, embedded memroy controllers and more instructions per cycle, but it is from the internal structure very close to an improved 68060 design.

On the PowerPC side.

The 603 and 604 have a similar name but are two very different designs.

The 750 (called G3) is basically an evolved 603.

While the 7400 (called G4) is an evolution of the 604.


posts 68page  1 2 3 4