Oepabakkes,
Quote:

Well, I sorta guessed that, but I thought it might me worth a gamble, it wasn't much more expensive than new stuff anyway, and I could find a place for the old stuff, so no great loss.

oh well, live and learn.

MDD 1.67 (OC)
Ram Disk:> cpu
System: PowerMac3,6
Machine: 0x1
PageSize: 4096 Bytes
CPUCount: 2
CPU0: 7445/7455 (G4) Version 32769 Revision 771
CPU0: CPUClock 1666666660 BusClock 166637218
CPU0: FPU Unit exists
CPU0: Altivec Unit exists
CPU0: Performance Monitor Unit exists
CPU0: DataStream Unit exists
CPU0: ReservationSize 32
CPU0: Ticks 4
CPU0: CacheL1Type 0x0
CPU0: CacheL1Flags 0x1000F
CPU0: ICacheL1Size 32768
CPU0: ICacheL1Lines 1024
CPU0: ICacheL1LineSize 32
CPU0: DCacheL1Size 32768
CPU0: DCacheL1Lines 1024
CPU0: DCacheL1LineSize 32
CPU0: CacheL2Type 0x0
CPU0: CacheL2Flags 0x1001D
CPU0: ICacheL2Size 262144
CPU0: ICacheL2Lines 0
CPU0: ICacheL2LineSize 0
CPU0: DCacheL2Size 262144
CPU0: DCacheL2Lines 0
CPU0: DCacheL2LineSize 0
CPU0: CacheL3Type 0x0
CPU0: CacheL3Flags 0x1001D
CPU0: ICacheL3Size 2097152
CPU0: ICacheL3Lines 0
CPU0: ICacheL3LineSize 0
CPU0: DCacheL3Size 2097152
CPU0: DCacheL3Lines 0
CPU0: DCacheL3LineSize 0
CPU1: 7445/7455 (G4) Version 32769 Revision 771
CPU1: CPUClock 1666666660 BusClock 166637218
CPU1: FPU Unit exists
CPU1: Altivec Unit exists
CPU1: Performance Monitor Unit exists
CPU1: DataStream Unit exists
CPU1: ReservationSize 32
CPU1: Ticks 4
CPU1: CacheL1Type 0x0
CPU1: CacheL1Flags 0x1000F
CPU1: ICacheL1Size 32768
CPU1: ICacheL1Lines 1024
CPU1: ICacheL1LineSize 32
CPU1: DCacheL1Size 32768
CPU1: DCacheL1Lines 1024
CPU1: DCacheL1LineSize 32
CPU1: CacheL2Type 0x0
CPU1: CacheL2Flags 0x1001D
CPU1: ICacheL2Size 262144
CPU1: ICacheL2Lines 0
CPU1: ICacheL2LineSize 0
CPU1: DCacheL2Size 262144
CPU1: DCacheL2Lines 0
CPU1: DCacheL2LineSize 0
CPU1: CacheL3Type 0x0
CPU1: CacheL3Flags 0x1001D
CPU1: ICacheL3Size 2097152
CPU1: ICacheL3Lines 0
CPU1: ICacheL3LineSize 0
CPU1: DCacheL3Size 2097152
CPU1: DCacheL3Lines 0
CPU1: DCacheL3LineSize 0

Ram Disk:> memtest disable
Allocated memory at 0x2578F140 -> 0x5C87FB3F, 923732480 bytes
Writing and Verifying 0x00000000...
Write 765 MB/sec. Verify 405 MB/sec.
Writing and Verifying 0xFFFFFFFF... Write 767 MB/sec. Verify 405 MB/sec.
Memory is ok!
Ram Disk:> Ram Disk:> memtest disable
Allocated memory at 0x2578F140 -> 0x5C87FB3F, 923732480 bytes
Writing and Verifying 0x00000000...
Write 767 MB/sec. Verify 405 MB/sec.
Writing and Verifying 0xFFFFFFFF... Write 767 MB/sec. Verify 405 MB/sec.
Memory is ok!

It seems nobody has posted Efika results until now...
System: bplan,EFIKA5K2
Machine: 0x1
PageSize: 4096 Bytes
CPUCount: 1
CPU0: 5200 (G2-LE) Version 32898 Revision 8212
CPU0: CPUClock 396000000 BusClock 132000000
CPU0: FPU Unit exists
CPU0: ReservationSize 4
CPU0: Ticks 4
CPU0: CacheL1Type 0x0
CPU0: CacheL1Flags 0x1000F
CPU0: ICacheL1Size 16384
CPU0: ICacheL1Lines 512
CPU0: ICacheL1LineSize 32
CPU0: DCacheL1Size 16384
CPU0: DCacheL1Lines 512
CPU0: DCacheL1LineSize 32

Allocated memory at 0x2312A280 -> 0x25E0557F, 47035136 bytes
Writing and Verifying 0x00000000... Write 142 MB/sec. Verify 131 MB/sec.
Writing and Verifying 0xFFFFFFFF... Write 142 MB/sec. Verify 131 MB/sec.
Writing and Verifying 0xFFFF0000... Write 142 MB/sec. Verify 131 MB/sec.
Writing and Verifying 0x0000FFFF... Write 142 MB/sec. Verify 131 MB/sec.
Writing and Verifying 0xFF00FF00... Write 142 MB/sec. Verify 131 MB/sec.
Writing and Verifying 0x00FF00FF... Write 142 MB/sec. Verify 131 MB/sec.
Writing and Verifying 0xF0F0F0F0... Write 142 MB/sec. Verify 131 MB/sec.
Writing and Verifying 0x0F0F0F0F... Write 142 MB/sec. Verify 131 MB/sec.
Writing and Verifying 0x55555555... Write 142 MB/sec. Verify 131 MB/sec.
Writing and Verifying 0xAAAAAAAA... Write 142 MB/sec. Verify 131 MB/sec.
Writing and Verifying 0xAAAA5555... Write 142 MB/sec. Verify 131 MB/sec.
Writing and Verifying 0x5555AAAA... Write 142 MB/sec. Verify 131 MB/sec.
Writing and Verifying 0xAA55AA55... Write 142 MB/sec. Verify 131 MB/sec.
Writing and Verifying 0x55AA55AA... Write 142 MB/sec. Verify 131 MB/sec.
Writing and Verifying 0xA5A5A5A5... Write 142 MB/sec. Verify 131 MB/sec.
Writing and Verifying 0x5A5A5A5A... Write 142 MB/sec. Verify 131 MB/sec.
Memory is ok!


[ Edited by analogkid 02.09.2011 - 15:22 ]

> System: bplan,EFIKA5K2 [...]
> Write 142 MB/sec. Verify 131 MB/sec.

This made me think of those comments by a MorphOS Team member:

http://www.amiga-news.de/de/news/comments/262773.html
http://www.amiga-news.de/de/news/comments/262821.html

;-)

@Andreas Wolf

Maybe I don't get your point, but in my eyes there's no contradiction between the results and Geit's statement.

> in my eyes there's no contradiction between the results and Geit's statement.

Well, I wouldn't say that the figures you just posted back up the statements that the Efika's "Speicher ist so verdammt schnell" and that the Efika's "RAM-Speed haut da gewaltig rein".
And as for "den G3 600 steckt es locker weg" I'm not so sure as well:

https://morph.zone/modules/newbb_plus/viewtopic.php?forum=11&topic_id=7474&start=6

As you can see, your Efika is faster at writing but slower at reading. And that's compared to a machine with a crippled northbridge (assuming the Peg2 G3 has better memory performance than Peg1 G3).

@Andreas Wolf:

Well, i think it must be seen in relation to the whole system performance. And a Peg2/G3 isn't faster regarding speed of memory, I can remember the memtest results some years ago (it was my first MorphOS 2.x system).

> i think it must be seen in relation to the whole system performance.

That's exactly my point. The statements I quoted don't sound relative at all, also not in their original context.

> a Peg2/G3 isn't faster regarding speed of memory

Fair enough. As I see it, your figures prove that the Efika does not "den G3 600 steckt es locker weg", even when Peg1 G3 and Peg2 G3 are on par in terms of memory performance.

> That's exactly my point. The statements I quoted don't sound relative at all, also not in their original context.

I'm not supporting your opinion. There may be some room of speculation in Geits statement, but as an Efika user I unterstand and can stress his statement, namely the relativly fast memory speed. It should be clear that the Efika's memory speed can't compete with the memory speed of a Mac mini or Powermac G4 MDD. The limiting factor is the CPU, because someone/something has to copy the content from/to memory...
And Geit is right, if he stresses the better memory timing of the Efika, in comparison to a Pegasos board (no matter which one), which has a quite conservative memory timing.

> Fair enough. As I see it, your figures prove that the Efika does not "den G3 600 steckt es locker weg", even when Peg1 G3 and Peg2 G3 are on par in terms of memory performance.

Maybe you are right in absolute numbers for some statistics, but the user experience gives him (and me, supporting his opinion) right.

> I unterstand and can stress his statement, namely the relativly fast memory speed.

I don't support your opinion that there's anything relative in the statements I quoted.

> The limiting factor is the CPU

Exactly. That's why the Efika's memory speed is as low as "Write 142 MB/sec. Verify 131 MB/sec", which I don't consider "verdammt schnell" and "haut da gewaltig rein", especially with a SoC, i.e. CPU and memory controller in the same chip.

> Geit is right, if he stresses the better memory timing of the Efika, in comparison
> to a Pegasos board

I don't know about timing, but I know that your figures prove that Efika does not "den G3 600 steckt es locker weg", no matter the timing.

> Maybe you are right in absolute numbers for some statistics

The "some statistics" I refer to are the numbers from this very thread. Of course I can't assure that the members who posted them didn't manipulate them. But then I don't see a reason why.

> the user experience gives him (and me, supporting his opinion) right.

I hope you don't mind that I trust the hard and cold numbers more than the "experience" of you both.

> I don't support your opinion that there's anything relative in the statements I quoted.

Well, you are free to support what you want.

> The "some statistics" I refer to are the numbers from this very thread. Of course I can't assure that the members who posted them didn't manipulate them. But then I don't see a reason why.

I'm sure no one manipulated the numbers from this thread, including me (except you'd call me a liar, out of the reason I can stress Geits arguments pro Efika...).

> I hope you don't mind that I trust the hard and cold numbers more than the "experience" of you both.

Well, you are free to trust what and who you want.

For the sake of my and the audiences' nerves I would like to end this off-topic discussion.

> you are free to support what you want.

Agreed.

> I'm sure no one manipulated the numbers from this thread

Agreed.

> except you'd call me a liar, out of the reason I can stress Geits arguments
> pro Efika...

It wouldn't make sense to lower the Efika numbers in order to support his statement, now would it? ;-)

> you are free to trust what and who you want.

Agreed.

> I would like to end this off-topic discussion.

I won't prevent you from doing just that.

Today I was testing a few CPU upgrades in a Digital Audio Powermac that I use for testing. I decided to try my 7448 @ 2GHz to see whether memory performance was any different to when it is installed in my Quicksilver 2002. I got the following results:

Ram Disk:> memtest disable
Allocated memory at 0x26833980 -> 0x6C4CD8BF, 1170841408 bytes
Writing and Verifying 0x00000000... Write 757 MB/sec. Verify 258 MB/sec.
Writing and Verifying 0xFFFFFFFF... Write 764 MB/sec. Verify 258 MB/sec.
Memory is ok!

A recap of my Quicksilver 2002 reults (same memory used):

Allocated memory at 0x2CB04B40 -> 0x5C9137BF, 803269760 bytes
Writing and Verifying 0x00000000...
Write 308 MB/sec. Verify 381 MB/sec.
Writing and Verifying 0xFFFFFFFF... Write 308 MB/sec. Verify 381 MB/sec.
Memory is ok!

Quite a difference...
Based on this result, maybe Digital Audio is the better choice

Ruud: no mate, my overclocked MDD 1.67Ghz is still faster but not stable I finded out recently.. Have to try with 1.58..

pampers,
Quote:

Sure MDD memory performance is still better (so is mac mini for that matter). I meant that maybe Digital Audio is the better choice for my MaxPower 7448 @ 2GHz than a Quicksilver. My hardware skills aren't upto hacking it to work in an MDD machine ;)

[ Edited by Ruud 21.01.2012 - 19:12 ]

@ruud

Not sure wether it's really the mobo that makes the difference here, cos with my QuickSilver+2x1.8GHz Sonnet I get 563 MB/sec for writing.

Even more interesting are the 383 MB/sec on verify, which slightly better than my Mini (376).

Questionable, the verify times stink.

I got slightly better memtest results now, with MorphOS 3.0 . Maybe because I have 3 similar (512Mb) sticks in my machine ?

stephen_robinson,
Quote:

That was with 2.7, with 3.0 on the same machine

Ram Disk:> memtest disable
Allocated memory at 0x246F7440 -> 0x5C8716BF, 941073024 bytes
Writing and Verifying 0x00000000... Write 782 MB/sec. Verify 377 MB/sec.
Writing and Verifying 0xFFFFFFFF... Write 783 MB/sec. Verify 377 MB/sec.
Memory is ok!

[ Edited by stephen_robinson 09.06.2012 - 15:18 ]

Powerbook 5,8

Ram Disk:> Ram Disk:> memtest disable
Allocated memory at 0x2D5EE900 -> 0x7F963B7F, 1379357312 bytes
Writing and Verifying 0x00000000...
Write 677 MB/sec. Verify 335 MB/sec.
Writing and Verifying 0xFFFFFFFF... Write 677 MB/sec. Verify 335 MB/sec.
Memory is ok!

Powerbook 5,6:
write 691 MB/s verify 351 MB/s

Ram Disk:> cpu
System: PowerBook5,9
Machine: 0x1
PageSize: 4096 Bytes
CPUCount: 1
CPU0: 7447A (G4) Version 32771 Revision 261
CPU0: CPUClock 1666666660 BusClock 166666666
CPU0: FPU Unit exists
CPU0: Altivec Unit exists
CPU0: Performance Monitor Unit exists
CPU0: DataStream Unit exists
CPU0: ReservationSize 32
CPU0: Ticks 20
CPU0: CacheL1Type 0x0
CPU0: CacheL1Flags 0x1000F
CPU0: ICacheL1Size 32768
CPU0: ICacheL1Lines 1024
CPU0: ICacheL1LineSize 32
CPU0: DCacheL1Size 32768
CPU0: DCacheL1Lines 1024
CPU0: DCacheL1LineSize 32
CPU0: CacheL2Type 0x0
CPU0: CacheL2Flags 0x1001D
CPU0: ICacheL2Size 524288
CPU0: ICacheL2Lines 0
CPU0: ICacheL2LineSize 0
CPU0: DCacheL2Size 524288
CPU0: DCacheL2Lines 0
CPU0: DCacheL2LineSize 0
Ram Disk:> memtest disable
Allocated memory at 0x23E82480 -> 0x7F8A1E7F, 1537341952 bytes
Writing and Verifying 0x00000000... Write 679 MB/sec. Verify 335 MB/sec.
Writing and Verifying 0xFFFFFFFF... Write 677 MB/sec. Verify 335 MB/sec.
Memory is ok!

Zylesea,
Quote:

So the DDR1 model does have faster memory than the DDR2 model that followed, hmmm..

> So the DDR1 model does have faster memory than the DDR2 model that
> followed, hmmm..

Yes, and this has already been discussed here on MorphZone.org:

https://morph.zone/modules/newbb_plus/viewtopic.php?topic_id=8264&forum=11&start=8

my powerbook 1.67 (5.6) did

Ram Disk:> cpu
System: PowerBook5,6
Machine: 0x1
PageSize: 4096 Bytes
CPUCount: 1
CPU0: 7447A (G4) Version 32771 Revision 258
CPU0: CPUClock 1666666660 BusClock 166483989
CPU0: FPU Unit exists
CPU0: Altivec Unit exists
CPU0: Performance Monitor Unit exists
CPU0: DataStream Unit exists
CPU0: ReservationSize 32
CPU0: Ticks 9
CPU0: CacheL1Type 0x0
CPU0: CacheL1Flags 0x1000F
CPU0: ICacheL1Size 32768
CPU0: ICacheL1Lines 1024
CPU0: ICacheL1LineSize 32
CPU0: DCacheL1Size 32768
CPU0: DCacheL1Lines 1024
CPU0: DCacheL1LineSize 32
CPU0: CacheL2Type 0x0
CPU0: CacheL2Flags 0x1001D
CPU0: ICacheL2Size 524288
CPU0: ICacheL2Lines 0
CPU0: ICacheL2LineSize 0
CPU0: DCacheL2Size 524288
CPU0: DCacheL2Lines 0
CPU0: DCacheL2LineSize 0
Ram Disk:> memtest extensive disable
Allocated memory at 0x281622C0 -> 0x5C8BC33F, 880124032 bytes
Writing and Verifying 0x00000000...
Write 693 MB/sec. Verify 351 MB/sec.
Writing and Verifying 0xFFFFFFFF... Write 692 MB/sec. Verify 343 MB/sec.
Writing and Verifying 0xFFFF0000... Write 838 MB/sec. Verify 409 MB/sec.
Writing and Verifying 0x0000FFFF... Write 838 MB/sec. Verify 407 MB/sec.
Writing and Verifying 0xFF00FF00... Write 838 MB/sec. Verify 409 MB/sec.
Writing and Verifying 0x00FF00FF... Write 838 MB/sec. Verify 396 MB/sec.
Writing and Verifying 0xF0F0F0F0... Write 838 MB/sec. Verify 409 MB/sec.
Writing and Verifying 0x0F0F0F0F... Write 838 MB/sec. Verify 399 MB/sec.
Writing and Verifying 0x55555555... Write 838 MB/sec. Verify 409 MB/sec.
Writing and Verifying 0xAAAAAAAA... Write 838 MB/sec. Verify 403 MB/sec.
Writing and Verifying 0xAAAA5555... Write 838 MB/sec. Verify 409 MB/sec.
Writing and Verifying 0x5555AAAA... Write 838 MB/sec. Verify 407 MB/sec.
Writing and Verifying 0xAA55AA55... Write 838 MB/sec. Verify 409 MB/sec.
Writing and Verifying 0x55AA55AA... Write 838 MB/sec. Verify 408 MB/sec.
Writing and Verifying 0xA5A5A5A5... Write 838 MB/sec. Verify 409 MB/sec.
Writing and Verifying 0x5A5A5A5A... Write 838 MB/sec. Verify 404 MB/sec.
Memory is ok!

bye, MarK.

[ Edited by MarK 12.06.2012 - 09:32 ]