Mac OS 9 Lives
Mac OS 9 Discussion => Hardware => Topic started by: Knezzen on March 10, 2017, 04:48:29 PM
-
Just browsed through the old Geekbench archives of benchmarked MDD's.
I know that Geekbench is OSX software, but I look at it as a somewhat accurate benchmark application for the bare hardware since it's near impossible to get an accurate benchmark from anything Mac OS 9 based.
http://browser.primatelabs.com/geekbench2/271199
This MDD is scoring 2253 points and it all sort of looks like it's coming from an insane CPU integer score.
In contrast to this, my dual 1.67ghz Sonnet MDX (7447A) powered MDD scores around 1500 points and the 1.8ghz Sonnet's seems to score around 1700 points. This looks like a regular Apple 1.42ghz 7455 unit overclocked to 1.67ghz. Something must have been wrong when the benchmarking was made, right?
Any thoughts? :)
-
it almost looks like a typo :)
-
it almost looks like a typo :)
Can't be. The database only accepted data sent directly from Geekbench. So it's nothing that gets typed in place :P
-
This was uploaded by myself. ;D
Even though I don't trust Geekbench results ...
This machine is powerful and fast.
I think that this results can be surpassed if I do the test again, with the Ti4600 and the SSD connected to the Seritek 64 bits card.
-
Really strange. I'm getting 1440 points with my machine. Dual 1.67ghz 7447A's (Sonnet MDX), 128gb SSD connected to a 64bit Seritek, GeForce Ti4600 (OEM Apple card). Which version of GeekBench where you using there? Could you do a run with Xbench for comparison? I'm just curious if the result is real or if it's GeekBench showing an error.
You can find Xbench here: http://www.xbench.com/
Fresh results from my MDD:
(http://macos9lives.com/smforum/index.php?action=dlattach;topic=3718.0;attach=4087)
-
but it is strange that only one thing is affected. integer vs float should not have anything to do with cache size or something like that
-
Could you do a run with Xbench for comparison? I'm just curious if the result is real or if it's GeekBench showing an error.
You can find Xbench here: http://www.xbench.com/
It will take me a few days... I have to set up a PPC OS X system. Believe me or not I have none.
I prefer the MacBench 3.0 results: ;D
(http://macos9lives.com/smforum/index.php?action=dlattach;topic=944.0;attach=264)
-
I prefer the MacBench 3.0 results: ;D
I'll use MacBench 3.0 then. Can you post the version you're using, so I know we're using the same version? :)
For comparison; this is my MDD when it had a mechanical HD connected to the internal connector, Radeon 9000 64mb and a dual 1.25ghz module overclocked to 1.5ghz.
http://browser.primatelabs.com/geekbench2/675473
-
I'll use MacBench 3.0 then. Can you post the version you're using, so I know we're using the same version? :)
MacBench 3.0 :)
-
Was thinking more like your actual MacBench install files, so we have the same actual install. But otherwise, will the one on Macintosh Garden work?
-
But otherwise, will the one on Macintosh Garden work?
Yes, I guess ... :)
-
Just browsed through the old Geekbench archives of benchmarked MDD's.
I know that Geekbench is OSX software, but I look at it as a somewhat accurate benchmark application for the bare hardware since it's near impossible to get an accurate benchmark from anything Mac OS 9 based.
http://browser.primatelabs.com/geekbench2/271199
This MDD is scoring 2253 points and it all sort of looks like it's coming from an insane CPU integer score.
In contrast to this, my dual 1.67ghz Sonnet MDX (7447A) powered MDD scores around 1500 points and the 1.8ghz Sonnet's seems to score around 1700 points. This looks like a regular Apple 1.42ghz 7455 unit overclocked to 1.67ghz. Something must have been wrong when the benchmarking was made, right?
Any thoughts? :)
it almost looks like a typo :)
This was uploaded by myself. ;D
LOL
-
9 world is getting smaller and smaller.
-
9 world is getting smaller and smaller.
Indeed :)
-
Just finished a run of MacBench 3.0. I didn't see any reason why your GeekBench results are that much higher then mine. I think there must be a bug in the version you where running at the time. The only way to find out is running the benchmark using the same version of GeekBench.
Can you explain my rubbish disk performance compared to yours, MacTron?
I'm using a KingSpec SSD connected to a 64bit SeriTek card, so it must be the SSD (I hope).
Does it matter which port on the card the SSD is connected to?
Do you know if it matters which PCI slot the card is inserted to? It shouldn't, but perhaps theres some "black magic" somewhere ;).
Another thing. You're scoring 727 points on "Publishing Graphics Mix" but my machine only manages 265 points even though we have the same type of card in our machines. Why is this?
(http://macos9lives.com/smforum/index.php?action=dlattach;topic=3718.0;attach=4089)
-
You're scoring 727 points on "Publishing Graphics Mix" but my machine only manages 265 points even though we have the same type of card in our machines. Why is this?
he has a special version of COM port installed which renders coregraphics stuff right on a casette tape at 2x speed.
-
LOL
-
note
my new
quote
-
Can you explain my rubbish disk performance compared to yours, MacTron?
I'm using a KingSpec SSD connected to a 64bit SeriTek card, so it must be the SSD (I hope).
Does it matter which port on the card the SSD is connected to?
Another thing. You're scoring 727 points on "Publishing Graphics Mix" but my machine only manages 265 points even though we have the same type of card in our machines. Why is this?
I don't know what's happen in to your system... but you should test the SSD with QuickBench, because it can badly affect to several test.
-
Can you explain my rubbish disk performance compared to yours, MacTron?
I'm using a KingSpec SSD connected to a 64bit SeriTek card, so it must be the SSD (I hope).
Does it matter which port on the card the SSD is connected to?
Another thing. You're scoring 727 points on "Publishing Graphics Mix" but my machine only manages 265 points even though we have the same type of card in our machines. Why is this?
I don't know what's happen in to your system... but you should test the SSD with QuickBench, because it can badly affect to several test.
I have done this before, see this post: http://macos9lives.com/smforum/index.php/topic,306.msg23522.html#msg23522
Im more curious to your insane graphics score. We have the exact same card, yet you get A LOT more points than me.
-
You can try Let1kwindowsbloom, it's useful to show a in excess bloated system folder ... and its dramatic repercussion in graphic and System responsiveness.
-
I installed the Universal Mac OS 9.2.2 from here today and it did the trick. Got more or less spot on the same graphics benchmark score in MacBench as you MacTron! Was using the "os9general.dmg" before. Going to go with the universal install on all machines now. The speed increase is very noticable.
Haven't really paid any attention to this before, thinking "how much of a difference can it make?". The answer is "a huge difference" ;) . Thanks for the great work guys!
-
Thanks for posting back Knezzen.
-
I installed the Universal Mac OS 9.2.2 from here today and it did the trick. Got more or less spot on the same graphics benchmark score in MacBench as you MacTron! Was using the "os9general.dmg" before. Going to go with the universal install on all machines now. The speed increase is very noticable.
Haven't really paid any attention to this before, thinking "how much of a difference can it make?". The answer is "a huge difference" ;) . Thanks for the great work guys!
I don't think that's the real cause, as "Mac OS 9.2.2 Universal Install" and "os9general.dmg" are pretty the same. ( Mac OS ROM Version 10.2.1, Mac OS CPU Software 5.9 from 04.2003 )
... maybe you have some issues in your old system folder ...
In fact I have my own system folder used across all my G4s. Based on "os9general.dmg".
-
I don't think that's the real cause, as "Mac OS 9.2.2 Universal Install" and "os9general.dmg" are pretty the same. ( Mac OS ROM Version 10.2.1, Mac OS CPU Software 5.9 from 04.2003 )
... maybe you have some issues in your old system folder ...
In fact I have my own system folder used across all my G4s. Based on "os9general.dmg".
I have the same extensions activated in both system folders. Everything is the same.
You're probably right, but I don't know where to start looking or if it's worth it at all since everything works like it should now. See attached picture :)
(http://macos9lives.com/smforum/index.php?action=dlattach;topic=3718.0;attach=4095)
-
The Processor and Floating point are better than I could expect. The Ti4600 performs incredible well as it could be expected. But otherwise the "Disk Mix" and "Publishing Disk Mix" is a bit low... it's weird.
-
The Processor and Floating point are better than I could expect. The Ti4600 performs incredible well as it could be expected. But otherwise the "Disk Mix" and "Publishing Disk Mix" is a bit low... it's weird.
Probably due to my KingSpec SSD. Should put in a SSD from a "good" brand in it and see what happens to the benchmarks.