debian hardinfo benchmarks

[KimmoK]

T4 chips are clocked up to 1.8Ghz.
But they seem to have downgraded e6500 info to 2Ghz.
http://www.freescale.com/files/32bit/doc/brochure/PWRARBYNDBITSQIG.pdf

The old info:
http://media.freescale.com/phoenix.zhtml?c=196520&p=irol-newsArticle&ID=1576370

[drHirudo]

Because this is the truth, so people may know it.

Hail for Piru - the truth-bringer. The same Piru who brought the lie about the Amikit breaking laws by avoiding European warranty periods.

Oh, well. Whatever.

Link to the thread

[Iggy]

T4 chips are clocked up to 1.8Ghz.
But they seem to have downgraded e6500 info to 2Ghz.
http://www.freescale.com/files/32bit/doc/brochure/PWRARBYNDBITSQIG.pdf

The old info:
http://media.freescale.com/phoenix.zhtml?c=196520&p=irol-newsArticle&ID=1576370

Yes, the T4 family maxes out at 1.8 GHz and they haven't announced any T1, T2, or T3 products yet.
So, while Freescale claims the e6500 core can run at up to 2 GHz, they haven't pushed it to that yet.

I'm not sure I'm interested in AMP processors outside of the T4280.
That processor offers the most SerDes lanes (other processors in this family have less).
This will allow it to support better expansion options (like 16X PCIe cards).

[minator]

I'm still surprised at the fairly low results.  Though it's interesting how close it is to the higher clocked G5 running the same benchmark.

There could be a number of reason for this:

System is set up wrong, but in OS 4 and Linux seems unlikely (unless the same people did both).


The low power has involved compromises that impact its performance - This is quite likely, you don't get low power for free.

The PA6T is "fragile" performance wise - A distinct possibility, this is not exactly rare even at the high end. The G5 was behind the x86s at general purpose stuff but ahead on heavy maths.  Could be something similar but more pronounced - its weak on general purpose but the FFT score is very good.

One of the modes is faster than the other - has anyone compared 32 bit to 64 bit Linux to see if it makes a difference?

Software is compiled for the wrong target - The PA6T was in part designed to have similar characteristics to the G5, to the point that one of the G5's weak points was copied (2 cycle integer instructions).  Code compiled for the G5 should run better than code for the G4.

[jorkany]

the truth would be nice eh..see post 20.

@ Piru

I know you don't want to troll this way dude so tell 'the truth' on this thread and change those graphs to refelct that the X1000 is only operating on one core.

How many times must you be told: all of the machines in the graph are running on one core. If any of them were running on two cores you might have some kind of point, but they aren't, none of them. Just because you don't understand this doesn't mean Piru is "trolling".

[Iggy]

How many times must you be told: all of the machines in the graph are running on one core. If any of them were running on two cores you might have some kind of point, but they aren't, none of them. Just because you don't understand this doesn't mean Piru is "trolling".

I'd like to see Piru post dual core benchmarks.
How about a G4 FW800 Power mac, a G5 Power mac, and the X1000.

[takemehomegrandma]

The low power has involved compromises that impact its performance - This is quite likely, you don't get low power for free.

Of course it has, and IIRC the key goal for the PA6T was low power (Watts), not high performance. AFAIK they targeted it for Apple to use in laptops post G4, not to compete with the G5 desktops. That never happened though, as we all know, it only became a "could have been" parenthesis in computer history (was it even really launched in a final state, or only as samples?)...

[takemehomegrandma]

I'd like to see Piru post dual core benchmarks.

Why don't *you* do it, you are one of those having a dual CPU Mac, aren't you? You might have to choose a different test suite though?

Dual core benchmarks are pointless in an Amiga context though, so why bother? Amiga won't ever be SMP without breaking the "Amiga" in it. I would vote "aye" for doing that in future MorphOS, but my view of the OS4 community is that it's largely built on the "we are based on teh tru sources", "we are teh reel!!1!" and other mumbo jumbo statements, so I don't think it will go down just as easily for them. And of course, if you are going to cut the cord, break from the past, and start with a clean slate, why on earth would anyone be stupid enough to do it on an obscure, backwater PPC platform? Doesn't compute...

[jorkany]

Dual core benchmarks are pointless in an Amiga context though, so why bother? Amiga won't ever be SMP without breaking the "Amiga" in it. I would vote "aye" for doing that in future MorphOS, but my view of the OS4 community is that it's largely built on the "we are based on teh tru sources", "we are teh reel!!1!" and other mumbo jumbo statements, so I don't think it will go down just as easily for them. And of course, if you are going to cut the cord, break from the past, and start with a clean slate, why on earth would anyone be stupid enough to do it on an obscure, backwater PPC platform? Doesn't compute...

If ALL the machines are benchmarked using two cores then that's a valid comparison, but what does anyone expect? ALL the machines will simply be faster, it isn't going to magically make the X1000 change position. But like you, I wonder what is the point of doing so if the X1000 is involved? The OS the X1000 is supposedly designed for doesn't have SMP or any other sort of multi-core support, so all it will do is expose OS4 users to precisely how little of the processing power of the X1000 is utilized.

Hey I know, maybe if the benchmark suites were redesigned to take advantage of the Power of X, the onboard Xena XMOS chip, then the scales will tip! Yeah that's the ticket!  :crazy smiley: :crazy smiley:

[klx300r]

How many times must you be told: all of the machines in the graph are running on one core. If any of them were running on two cores you might have some kind of point, but they aren't, none of them. Just because you don't understand this doesn't mean Piru is "trolling".

c'mon seriously now you do realize it is very misleading to not clarify on the graphs that the benchmarks in question only utilize ONE CPU core.  

As it stands those numbers are totally useless and Piru knows it unless you want to only compare the Samflex@800 and Apple product.

So let's do a 3D Blender benchmark and only use 1 core of the X1000 and then use the full potential of the CPU and compare the results under Debian hmmm I wonder which result would be better:rolleyes:

[Kesa]

Exactly. That's why i don't bother with benchmarks. Benchmarks are like thinktanks. Thinktanks are usually used when certain parties have an agenda and then use stats to prove it. Piru is just sugar coating the numbers in the benchmark to belittle the X1000 for the purpose of promoting Morphos. Sure, numbers don't lie - but people do.

[dammy]

Exactly. That's why i don't bother with benchmarks. Benchmarks are like thinktanks. Thinktanks are usually used when certain parties have an agenda and then use stats to prove it. Piru is just sugar coating the numbers in the benchmark to belittle the X1000 for the purpose of promoting Morphos. Sure, numbers don't lie - but people do.

I'm sure the benchmarks for the upcoming Galaxy SIII (European, not the Qualcomm American version) will provide some interest as well this month with it's quad core 1.5GHz A9 CPUs.   Wonder who the winner will be.

[klx300r]

Exactly. That's why i don't bother with benchmarks. Benchmarks are like thinktanks. Thinktanks are usually used when certain parties have an agenda and then use stats to prove it. Piru is just sugar coating the numbers in the benchmark to belittle the X1000 for the purpose of promoting Morphos. Sure, numbers don't lie - but people do.

yes we all know Piru's 'agenda' with this thread but I think he is usually fair with his facts and will change the info on the graphs to reflect the truth of the benchmarks.

[Iggy]

I'm sure the benchmarks for the upcoming Galaxy SIII (European, not the Qualcomm American version) will provide some interest as well this month with it's quad core 1.5GHz A9 CPUs.   Wonder who the winner will be.

Whoa!
You actually made a valid point AND you didn't make a CUSA comment.
I'm impressed.
Yeah a quad core A9 at 1.5 GHz would be pretty impressive.
I'd love to see future variants of MorphOS or AOS on that.

But we'll probably see AROS running on ARM before either of those two OS' gets ported.

[Piru]

For a real world example run a Blender demo on only one cpu and then run it with all the cpu's cores and then you will see real 'facts'

See Linux PowerPC Blender benchmark

[minator]

IIRC the key goal for the PA6T was low power (Watts), not high performance. AFAIK they targeted it for Apple to use in laptops post G4, not to compete with the G5 desktops.

Actually, it was designed to provide G5 type performance at much lower power - and if you look at the benchmarks of the X1000 and the G5 2.3GHz is is remarkably close on some of the benchmarks.