@Rose

Quote:

Are you scared of independet standard benchmarks which you cant rig?

the only benchmark that i know is my real hardware and not fluffy not useful trolling

_________________
I love Amiga and new hope by AmigaNG
A 500 + ; CDTV; CD32;
PowerMac G5 Quad 8GB,SSD,SSHD,7800gtx,Radeon R5 230 2GB;
MacBook Pro Retina I7 2.3ghz;
#nomorea-eoninmyhome

@tlosm

So basicly your "benchmarks" are pulled from your a**. And if I were you I would look what trolling means. Asking for proof for your claims hardly is trolling.

And while you always remeber to say that you are beta tester. Are you talking for A-eon. And are your "benchmarks" official standing of A-eon.

Last edited by Rose on 04-Jun-2016 at 12:18 AM.

@Rose

Quote:

r. Are you talking for A-eon.

no i tolking from me ... aeon didnt make me a present i had been pay for my machine ...
if you need benchmarks ask aeon for buy an x5000 like me and other and there you can make all the benchmarks thats you like.
... in different way ... take my info and my photos and dont reply because you cant.

_________________
I love Amiga and new hope by AmigaNG
A 500 + ; CDTV; CD32;
PowerMac G5 Quad 8GB,SSD,SSHD,7800gtx,Radeon R5 230 2GB;
MacBook Pro Retina I7 2.3ghz;
#nomorea-eoninmyhome

@tlosm

So honest benchmarks are out of question.

@Rose

Even if you have a point, you're still a dick.

@iggy

Nope, right term would be **** I think.

But you know as well as me that tsolm pulls X5000 performance from his ass.

This starts to feel too much like AmigaONE XE when there was nothing wrong with the motherboard.

I Guess honesty is too much to ask in Amiga scene.

So you buy his claim that X5000 is par with i5 in performance?



Mod Note: Foul Language removed

Last edited by _Steve_ on 28-Jun-2016 at 01:27 AM.
Last edited by Rose on 04-Jun-2016 at 02:22 AM.
Last edited by Rose on 04-Jun-2016 at 02:18 AM.
Last edited by Rose on 04-Jun-2016 at 01:51 AM.

@Aslak3

Of course there's different level of programming tasks and programmers. But some of people are always trying to say if something doesn't meet your personal expectations the authors must be morons. Certain programming tasks takes always certain amount of time no matter how smart you feel yourself.

_________________
Rock lobster bit me. My Workbench has always preferences. X1000 + AmigaOS4.1 FE
"Anyone can build a fast CPU. The trick is to build a fast system." -Seymour Cray

@Rose

Oh. Cool.
Yeah, its a rediculous claim.
I mean I like the cpus based on the e5500 core, but it wouldn't even come close.

And I remember the whole Artica S fiasco with people claiming it was fine and other people (like Bill Buck) insisting it was flawed.

The X5000 won't be perform anything like what he is suggesting, but it won't be flawed.
I am familar with the design firm. Its top notch.

Of course all A-eon boards are really high priced, and we should have switched ISAs awhile ago, but the fanatics will still rant you know?

BTW - I really like the way you dealt with my comment.

Jim

@Rose

Quote:

Are you scared of independet standard benchmarks which you cant rig?

Trust me, every benchmark can be rigged...

As of P5020 2 GHz performance (single core integer), it could be between 6 and 8 SpecInt2006 (my estimate based on known benchmarks and SpecInt2006 value for T4240 1.666 GHz).

@tlosm@Rose

Please, calm down, this thread slowly degrades to really nasty flamefest.

@pavlor

this is my last basilisk II result

https://scontent-mxp1-1.xx.fbcdn.net/v/t1.0-9/13254605_10206363423546033_5329338606939028498_n.jpg?oh=4e4412f95bfde6e21f3416bd8272cdef&oe=57CFF450

_________________
I love Amiga and new hope by AmigaNG
A 500 + ; CDTV; CD32;
PowerMac G5 Quad 8GB,SSD,SSHD,7800gtx,Radeon R5 230 2GB;
MacBook Pro Retina I7 2.3ghz;
#nomorea-eoninmyhome

@tlosm

So Basilik is running as an 100-150MHz 68040 on your X5000 ?
(edit: thought the Quadra 950 was 25MHz but it's 33MHz so make that a 130-200MHz 040)

Doesn't sound impressive considering that even a non-JIT UAE would reach such numbers on 10++ years old x86 HW (much more with JIT).

So either Basilik is superslow by design or the X1000 is about a decade behind in performance compared to top-notch x86.

Last edited by Kronos on 04-Jun-2016 at 09:46 AM.

_________________
- We don't need good ideas, we haven't run out on bad ones yet
- blame Canada

@Kronos

search and compare it with a quad g5 result
and compare it with a month ago result.. still in beta performace will increase for about 30% in basilisk in future.

about x86 and jit ... jit is enabled by default in x86 src and is disabled only in BE....
ps: jit is enabled from 10 years and more on x86

Last edited by tlosm on 04-Jun-2016 at 09:54 AM.

_________________
I love Amiga and new hope by AmigaNG
A 500 + ; CDTV; CD32;
PowerMac G5 Quad 8GB,SSD,SSHD,7800gtx,Radeon R5 230 2GB;
MacBook Pro Retina I7 2.3ghz;
#nomorea-eoninmyhome

@tlosm

Well Rose mentioned an i5 and comparing the X5000 to a current Intel is 100% bonkers.

If that was just a typo and a G5 was meant we are in somewhat realistic territories.


Wether JIT was availble on x86 for xx years isn't the point as atleast UAE allows switching it of making results to running on PPC comparable.


Still not impressed by the speed of that emulated 040

_________________
- We don't need good ideas, we haven't run out on bad ones yet
- blame Canada

@Kronos

The X5000 euae without jit gave me 42 mips , on G5 quad is 33 mips...
The jit is not optimized for 5020 and geve me result form 55 to 80 mips on sysinfo ... no realistic

_________________
I love Amiga and new hope by AmigaNG
A 500 + ; CDTV; CD32;
PowerMac G5 Quad 8GB,SSD,SSHD,7800gtx,Radeon R5 230 2GB;
MacBook Pro Retina I7 2.3ghz;
#nomorea-eoninmyhome

@tlosm

Quote:

The X5000 euae without jit gave me 42 mips , on G5 quad is 33 mips...

Faster RAM may be the deciding factor here, but somehow I don't see why emulating another system would be the correct benchmark, considering that native ones are available (mplayer, blender etc).

_________________
- We don't need good ideas, we haven't run out on bad ones yet
- blame Canada

@Kronos

Quote:

Faster RAM may be the deciding factor here, but somehow I don't see why emulating another system would be the correct benchmark, considering that native ones are available (mplayer, blender etc).

Although tlosm´s use of benchmarks is rather creatice (Piru´s style ), emulation of other systems is valid application benchmark - if you know what you do of course...

@Kronos

Belive me if i continue say what you had been see since today will be different in future ...
have finally a 64bit full OS and many many other fixes will rise all the benchmarks and results.
The system is too young and freescale engeeneer are continue fixing kernels and something else day by day

_________________
I love Amiga and new hope by AmigaNG
A 500 + ; CDTV; CD32;
PowerMac G5 Quad 8GB,SSD,SSHD,7800gtx,Radeon R5 230 2GB;
MacBook Pro Retina I7 2.3ghz;
#nomorea-eoninmyhome

@pavlor

Is it fair to compare e6500 and e5500 cores?
The e6500 performs slightly better.

Further, we are talking about cpus with drastically different clock speeds.

Sorry, I just cannot see either core coming close to an i5 (and I am highly biased toward the PPC).

Edit - OK, a typo eh?
Comparisons with the G5 definately seem off too.
Again, clock speed differences.

Benchmarks can at times produce questionable results.
I'd have to second Kronos' call for benchs based on real applications.

Last edited by iggy on 04-Jun-2016 at 12:50 PM.
Last edited by iggy on 04-Jun-2016 at 12:46 PM.

@Rose

Quote:

But you know as well as me that tsolm pulls X5000 performance from his ass.

Wow. I am impressed that his ass is PowerPC based.
I guess the days of truly "Embedded Computing" are upon us.

_________________
question=(2b||!(2b))

@iggy

Quote:

Is it fair to compare e6500 and e5500 cores?

Quote:

The e6500 performs slightly better.

I quoted 6-8 SpecInt2006 estimate for comparison with mainstream world.

Quote:

Sorry, I just cannot see either core coming close to an i5 (and I am highly biased toward the PPC).

Exactly.

Quote:

I'd have to second Kronos' call for benchs based on real applications.

Well, emulators are classic case of application benchmarks...