debian hardinfo benchmarks

[takemehomegrandma]

This is the peak performance OS4 system, sold as new in 2012, being beaten by a 2005 level Mac *laptop*. It seems the A1X1K can't play 1080p video without GPU acceleration (that's not even there anyway), which a 2005 level PowerMac can do with no problem. Well at least it's cheap. Oh wait, $3,000...

Pegasos 2's, Amigaone's, Apple HW etc from 2004 are still being used today (although *this is* running them on overtime). If this life span is applied to the Amigaone X1000, sold as new in 2012 with sub-2005 level performance, it means that this is what many OS4 users will be using in 2020. If there still is a OS4 by then, which I'm not entirely certain about, given the chosen path they are travelling.

Madness. But it's not like they weren't warned...

[takemehomegrandma]

The fastest OS4.x was Pegasos II. A1X1k should be faster than a G4 ~1.1Ghz (I believe).

The Pegasos 2 G4 was 1.0GHz (although easily overclocked).

Still, benchmarking in Debian is pointless since no one is gonna buy a similar unit to run solely linux on it

I think this Linux benchmarking is a response to some people dismissing *the previous* set of benchmarks made on OS4 (and MorphOS) as "pointless", since those benchmarks didn't test the *HW* in an identical way (the differencies in OS's would impact results too much) and OS4 in its current state couldn't make full use of the A1X1K hardware, etc, etc. Hence the Linux tests!

So no one could characterize it as "modern" by any means.

Of course it's not "modern", it's being stomped by 2005 level Mac's! :rant:

But this is what they market as the peak performance Amigaone, today, in 2012.

[takemehomegrandma]

@Forcie

As far as I can see it is just a similar thread to this one: http://amigaworld.net/modules/newbb/viewtopic.php?mode=viewtopic&topic_id=35671

But I guess that thread is just a troll thread telling us why x86 PC is superior to OS4 systems? I bet the SAM and X1000 owners posting their results really wanted to hurt and upset people by showing simple hardware facts about their systems. Right?

Never mind Kesa. Sometimes it's difficult to know whether he's "trolling or just stupid" but obviously he's kind of confused here. :lol:

- Doing benchmarks isn't something bad.
- Posting benchmarks online isn't something bad.
- Discussing these benchmarks isn't something bad either.

Of course not!

Trying to inflict some kind of selective taboo on civil discussions and measurable, enlightening test data (like Kesa just did above) is bad, however. And so is trying to sell a product by trying to hide information about its performance. Where were Kesa with his lectures about "agenda's" when some people threw stones at the MorphOS team for going the Mac route, mocking MorphOS's lack of "new" hardware? Well, it turned out that this "crappy old Mac HW" is just as good -or indeed even better- than the big Messiah computer those people put forward as the right way to go (at 1/20 of the cost), and *then* Kesa gets all upset! Which is kind of funny, since the whole benchmark initiative, as well as all the numbers (except the PowerMac), comes from the OS4 community, by *their* initiative! "Agenda?" :lol:

* Past MorphOS HW (Pegasos1/April, Pegasos2) was always better than OS4 HW (AmigaOne), and much cheaper.

* Current MorphOS HW (a whole flora of mainstream Mac machines in various shapes and forms) is better than current OS4 HW (Sam and A1X1K), and much cheaper.

* Future MorphOS HW (x86 or ARM, that's the question, maybe both?) will definitely be better than anything the OS4 people can put forward in batches of 30 units based on PPC, and much cheaper.

Once again, it turned out that the MorphOS team had the winning strategy. While still being tied to the PPC platform, they looked around themselves and noticed how the market was full of cheap mainstream HW that were more powerful than anything else on the PPC market, and they said to themselves: "Why don't we use that?", and so they did! The A1X1K is a horror-example of what you get when going in the opposite direction. But the thing is, so was the Sam460. And so was the Sam440 before that. Time and time again, it becomes evident that some people never learn! And no wonder, when all they seem wanting to do, is to put their head in the sand and forbid any kind of discussions on these subjects, and then push ahead with one crazy project after another! This is killing the OS4 platform (not that I really care, IMHO the OS4 project was completely redundant from Day 1, and the Amiga community would have been much better off without it)...

[takemehomegrandma]

That would be nice if that was followed in other threads as well, but trolls will be trolls.

+1

Couldn't agree more!

[takemehomegrandma]

e6500 should not be far behind the top of x86

:lol:

[takemehomegrandma]

Care to elaborate?
To me it seem you have not looked at e6500 specs?

Only briefly, since it's not relevant for anything I'm interested in. The e6500 is not a CPU, it's a technology, and I believe the new "AMP" deal will mean a great step-up for the QorIQ Communications Platform that Freescale is offering. It seems it will mean four times the performance from previous offerings, as well as energy saving features such as the "Night mode", where data path delivers packets to only as many cores as required to process them, while the rest of the cores enter low power mode (and wakes as traffic increases again). Energy saving features in networks has been gaining importance during the last half decade (or more), I don't think anyone will deny that, so this may be one of the key features of the AMP concept. The T4240/T4160 chips indeed have an impressive DMIPS/MHz figure for an embedded processor, and coupled with speeds up to 1.8GHz, this will mean much for today's/future network infrastructure that is more and more being complicated by protocols such as IPSec and SSL, which require more than just simple IP packet forwarding. Freescale claims it can forward 50Gbits/s, and no doubt will this mean powerful gateways, routers, switches, proxy server applications, network storage applications, etc. I think this will mean a new position for Freescale in the market!

(not far behind is same as a lot less than decade behind, higher performing than low end x86 dekstop today)

Uh, not going to try to pretend I understood what you just wrote there, you lost me somewhere prior to the comma. (Was that some kind of discrete mathematics? Or just a riddle?)

:lol:

[takemehomegrandma]

Hey, I'm a big PPC fan, but a 1.8GHz processor (even with 8 cores) isn't going to be a threat to an X86.

Of course not...

[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...

[takemehomegrandma]

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.

1. The subject of this thread is "debian hardinfo benchmarks" on various machines. It's a HW discussion.
2. The benchmark initiative started on AmigaWorld.net, by OS4 people. It's a HW discussion.
3. The only thing Piru did here, was to add his personal machine to the Graph. It's a HW discussion.
4. MorphOS was never being discussed, the thread was never about it, it's a HW discussion, and the *only ones* trying to make this thread a MorphOS thread is *YOU* (1, 2, 3, 4), and *SPIRANTHO*: 1

This is obvious to *everyone else*, just pointing it out to *you*, since you obviously haven't noticed...

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.

Shooting the messenger delivering info you don't approve of, is a classic (but foul) knee-jerk response to try to "correct" a situation you don't like. Of course it won't change any underlying facts, and in the end, *you* will be the one looking bad in others eyes...

klx300r:
"You're trolling"
Suggesting it's not the truth
"Misleading, useless and Piru knows it
"Agenda"

drHirudo:
Trying to discredit the messenger with external and  irrelevant stuff

...and now *YOU* just tried to picture Piru as a liar, when all he did was posting a graph with data that *anyone* can verify. He did nothing more, nothing else, but that's obviously enough for you to try to discredit him in this foul way. But I dare you: Now it's up to you to back up your claims. Prove your claims of Piru being a liar, or stand exposed as the little troll you are.

I have long suspected "Kesa" being just another of those "alter-ego accounts" with an agenda, registered in 2010 (suspicious there already) and already with strong views on a selection of issues, always popping in to push a certain agenda in certain threads. This suspicion is growing stronger by every post you make. It would be interesting to know the real identity behind "Kesa"...

[takemehomegrandma]

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.

The "G4"/e600 was never a bad performer compared to the "G5" on a clock by clock basis, in some areas it was definitely the winner even (but of course the G5 ran at a much higher clock and had other advantages)

From the many, many various benchmarks, on various OS's we have seen during the last months, I'd say that the PA6T is comparable to a G4 class CPU in *performance*, but is comparable to a G5 class CPU in it being *64-bit*. (The last part is of course pointless in an Amiga context)

[takemehomegrandma]

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'

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:

See Linux PowerPC Blender benchmark

:whack:

:roflmao:

[takemehomegrandma]

@ Piru

thanks for the link but you forgot to update the graphs in your first post ?

Maybe I misunderstood the processor part of the Hardinfo benchmark, but isn't it a CPU test (which in today reads: Core) by design? If you want a view of real life effects of single or multiple cores you would use other tests, like indeed the Blender benchmark?

btw, you also forgot to mention on your new thread that Blender on the X1000 doesn't have 3D acceleration yet so those numbers should be better when 3D on Debian PPC arrives

...which won't affect anything at all, the 3D acceleration is purely a GUI thing in Blender.

[takemehomegrandma]

Dude, I posted a link with proof in my post.

The only reason to why you made that post was to shoot the messenger by trying to discredit him. It was completely off topic. Everyone can see that. And everyone can see that the PA6T performs just the same, despite your post, so it had zero impact! :lol:

[takemehomegrandma]

grandma, for the 10'th time, Hardinfo is NOT MULTI THREADED

Wasn't that *exactly* what I just said? :confused: It was what I meant, anyway...

our friend Piru knows this and chooses to not correct the graphs for some 'strange' reason.

Given the above, everything *is already correct*, modifying the data would be tampering with the results! If you want to measure the CPU's SMP capabilities you would have to use other benchmarks, like blender (or numerous others)! It seems you don't understand the benchmark...?

as for Blender results you see the difference that using 1 thread as compared to both right?

Yes? :confused:

having 3D acceleration should slightly improve the render times as they definitely won't get worse

Since none of the benchmarks are using GPU for rendering (it's supposed to be a benchmark of the *CPU*), it won't get any better either...