Welcome, Guest. Please login or register.

Author Topic: dnetc benchmarks  (Read 38261 times)

Description:

0 Members and 1 Guest are viewing this topic.

Offline Tripitaka

  • Hero Member
  • *****
  • Join Date: Jun 2005
  • Posts: 1307
    • Show only replies by Tripitaka
    • http://acidapple.com
Re: dnetc benchmarks
« Reply #74 on: February 05, 2012, 12:54:37 PM »
Quote from: Piru;679283
Well, nothing new to see here, MorphOS still crushes AmigaOS4 in this department. In fact, the MPlayer video decoding benchmark deliberately excluded displaying the decoded video, as this would have seriously crippled the X1000 result (the Radeon HD driver doesn't yet support overlay).


True for now, but in theory the X1000 could take a brand new PCIE card providing someone writes the drivers for it.
Falling into a dark and red rage.
 

Offline takemehomegrandma

  • Hero Member
  • *****
  • Join Date: Oct 2002
  • Posts: 2990
    • Show only replies by takemehomegrandma
Re: dnetc benchmarks
« Reply #75 on: February 05, 2012, 01:25:26 PM »
Quote from: amigadave;679267
Trevor just said (not a direct quote, but I am guessing he said something like this) "to hell with how much it is going to cost, we need a new, more powerful Amiga(One)".


I can't really understand how you can applaud something like this. To me it's just insane.

Think of a country like Ethiopia, suffering from drought, crop failure and famine. People are dying by starvation in thousands. They need something to eat to survive and develop as a people. If not, they will die.

Someone comes up with a plan to make cheap meatballs that would provide decent meals to the masses in the short to mid term (the Moana project) while you potentially could try to work something different out for the long term (probably a migration of the people to greener pastures, like x86 or ARM, where the people can survive and develop further).  

But the Generals leading the country (Hyperion) decides that this is not a "worthy enough" solution for Ethiopia(TM), that Ethiopia(TM) is all about being different and better (in some weird, backwards way), that what's needed is a "special Ethiopian(TM) dish", and only this will save the Ethiopian people.

The people cheer, as they always do when the Generals have spoken (no matter what they say) at least the higher society AmigaWorld.net people, and in particular the Ethiopian Noble class at Amigans.net. Now the direction is set, the road is clear. Ehtiopia(TM) will rise once again and become a player on the global arena.

A couple of Ethiopian farmers, the Acube (those originally behind the "mass meatball solution"), develops a plan according to the Generals wishes. Instead of meatball (what's really "special" about meatballs anyway?) they decide to create a Gourmet Dish of Pork Chops (the Sam 440). The Generals cheer, the upper class society (AmigaWorld.net) cheer, the Nobles (Amigans.net) cheer even more. Finally the wait is over, finally Ethiopia(TM) has its special meal that will solve the famine and make the people grow back, both in volume and significance.

Some people however (quite a few actually), openly questions this strategy. They say "in times of famine, poverty and mass starvation, how on earth would the Ethiopians be helped by 'gourmet dishes' that only a few could afford?". But those people are aggressively fought by the Generals minions (the upper class society and nobles) on a broad scale. "Ethiopia has always been about the Gourmet way of eating" they say, but by aggressively using words like "Trolls" and "Negativity", often crying for the Secret Thought Police (moderators) to intervene; "How on earth could Ethiopia possibly rise to its former glory, when these trolls are allowed to spread negativity about our new Gourmet Dish?". And not seldom does the Secret Thought Police abide.

The problem for Ethiopia however, isn't "negativity". It's famine. What was needed was a nutritious dish with enough vitamins and proteins, that could be broadly accessible and easily obtainable. The "Gourmet Dish" that those Acube farmers provided, turned out to be pretty much on par with a carrot stew (the $129 Efika MX, actually even below that, because the Efika MX can actually do quite "meaty" things by clever integration of various on-chip "mini-dishes" that spices things up quite considerably, actually bringing it to a meatball level in many cases). But while the "Acube Gourmet Dish" in practice provides a plain carrot stew, it *costs* like a "Angus Beef Tenderloin" dish (Core-i7 gaming level computer), possibly more. This has the effect that it will neither provide enough nutrition for its consumers, nor will it have particularly many consumers in the first place. The famine continues, the people of Ethiopia continues to decrease in numbers.
 
After a while, the Acube farmers produces a new "Gourmet Dish", the "Pork Tenderloin" (Sam 460). This is a dish that, while it seems a bit more "luxurious" than the Pork Chops, it does have some rather strange oddities and quirks. Once again the debate about the Generals strategy arises, once again the Generals minions have to fight "negativity" and "trolls".

The "Pork Tenderloin" dish in most cases turns out to be slightly below the *older* of the meatball dishes out there (the Pegasos2), being "Gourmet" only in its price tag, that still puts it way out of reach for most people. The Ethiopian population continue to decrease.

Now one of the Generals himself (Ben Hermans) steps up to bring order to the famine situation. Since the General Strategy of course has been right all along, the problems could only be explained by the dishes not being luxurious enough for the Ethiopians taste! Together with the rich, upper class society member Trevor Dickinson, he comes up with the Über Gourmet dish "the special Beef Tenderloin". It is so special, that it actually has a "jewel" in the middle of the plate, made of glass, because "the first Ethiopian(TM) dishes was originally made with a custom jewel, Ethiopians simply loves the thought of a shiny custom jewel in the center of the dish, and a glass jewel will surely satisfy this need of shininess".

Once again the debate arose. Critics said "You can't make a Beef Tenderloin dish based on an old meatball recipe", and "there is no need for shiny jewels in the 21st century, it will only make things more expensive by making the dish a lot more complex, and it's not even really a jewel but a $20 piece of glass", and once again the Generals minions had to come out and fight the "trolls" spreading "negativity". They also seemed to have some kind of competition among themselves, inventing potential uses for the jewel, each idea seeming more narcotic inspired than the other, and almost all suggested a complete ignorance of what kind of jewel it actually was.

Now when it's released, it turns out to be a "Gourmet Dish" only in its price tag. It turns out to be slightly above the meatball dishes that has been around since half a decade ago (but being denied to the Ethiopian people by their Generals), while its price tag is equal to not one, not two, but three or even *four* "Angus Beef Tenderloin" (Core-i7 gaming level computer) dishes! And the jewel? Well, it just sits there in the middle of the plate, and nobody has no idea of how it will add to the taste or nutrition of the dish.

Will this make the Ethiopian population grow? Will it make Ethiopia(TM) more significant, and a player on the global arena? No of course not. The madness is obvious to everyone but the Generals, the Nobles, and the Upper Class Society members of Ethiopia. To everyone outside, both in Ethiopia and in the world as a whole, the madness behind the idea makes it a laughing stock. It can't even be called Ethiopia(TM), but is referred to as EthiopiaOne (or parasite marketing by the Generals using quotation marks: "Ethiopia").

The X1000 is madness, the Sam is madness, the whole mindset that OS4 "Amigans" has on expensive, custome made HW is madness, it's a shameful waste of resources and it will not bring the OS4 platform forward one single bit, rather the opposite.
MorphOS is Amiga done right! :)
 

Offline takemehomegrandma

  • Hero Member
  • *****
  • Join Date: Oct 2002
  • Posts: 2990
    • Show only replies by takemehomegrandma
Re: dnetc benchmarks
« Reply #76 on: February 05, 2012, 01:26:53 PM »
(The story continues: ;) )

Well there is another country, a neighbor to Ethiopia, a country called Kenya (MorphOS). This is a country very similar to Ethiopia, it has similar history, a similar background. It has a similar population, people enjoying the East African way of doing things, how East African stuff work, etc. These things may be frown upon by European and Northern Americans (Apple and Wintel) who indeed are at a completely different technological level than Africa. But the people living in Eastern Africa likes and appreciate what they have, for what it is, and they can always take a trip to Europe or USA if they want to, some (most perhaps) even works there, so they are very well aware of the differences.

Anyway, being neighbors, also Kenya suffered from the drought, crop failure and famine. They also have the same agricultural platform (the PPC) but the leaders in this country approached the problem in a completely different way. They thought "we need growth and development, but our agricultural platform makes this very difficult, what should we do?"

The answer was not to try to invent new gourmet dishes that only a few could afford and wouldn't really fill the stomach of those who could. No the answer was to look around to see what was already there for them to utilize.

It turned out that there were lots of possibilities to utilize various kinds of tasty and nutritious meatball dishes, that may be considered a bit out of date by European or US standards, but still a vast improvement to the East African meatballs (Pegasos2, A1) they had been living on until the crop failed and brought a starvation. They came up with a Mini meatball dish (Mac Mini) that was an extremely low footprint dish, but contained everything most people needs and wants, and could be measured to be up to 2x as powerful as the "old" meatball dishes. They came up with a big-plate meatball dish (Power Mac) that could be expandable, which can either be extremely low cost, or extremely performing (judged by meatball standards).  They also came up with an even more low-cost dish that actually contained everything you need, including the plate, knife and fork (the eMac) that while being a bit bulky, it still performed a lot better than the old meatballs, and at an *incredible* price, practically for free (sometimes that really *is* the price). And now they are about to introduced a flat kind of meatball, a hamburger, that you will actually be able to carry with you and eat wherever you want (Power Book).

In Kenya, there is no starvation, everyone who wants to live there (if only as a hobby) can have very cheap and easily obtainable meatball dishes of various kinds. But most importantly, the difference between Ethiopia and Kenya seems to be that the Ethiopian leaders focus very much on the agricultural process (PPC is very important) and trying to create new dishes based on this (new custom made motherboards based on various PPC CPU's), while the Kenya leaders is more focused on perfecting the recipe's for the meatball dishes that's mainstream and cheap (developing and optimizing the OS and SW for the existing Mac PPC HW). At least in short term; the MorphOS 3.0 will introduce the "hamburger" (the portable Power Book) and will also mean a major leap in features and performance. Then in a mid term perspective, the meatball recipe's will be even more improved and perfected. And in the long term perspective, they will re-evaluate the whole agricultural infrastructure, which might mean a migration to greener pastures (x86 or ARM), which will definitely mean new dishes altogether, like *real* Beef Tenderloin, made on Angus, (instead of carrot stew with a Beef Tenderloin price tag).

If you look at MorphOS's growth graph over at morphzone.org, you will see how the graph got visibly steeper after the Mac HW support. The growth rate (that was constant both before and after the Mac HW) *increased* visibly due to this. People are migrating from Ethiopia to Kenya, and this is a fact. Most certainly because they like the East African ways, while wanting cheap but great meatball dishes instead of what the Ethiopian Generals can offer them.

With this in perspective, a bit "amusing" to actually see someone enjoying the benefits of Kenya paying the price tag of 3-4 Core-i7 systems, for an Ethiopian "Meatballs Plus" dish, to be able to say "I am an Ethiopian(TM)".
MorphOS is Amiga done right! :)
 

Offline Duce

  • Off to greener pastures
  • Hero Member
  • *****
  • Join Date: Jul 2009
  • Posts: 1699
    • Show only replies by Duce
    • http://amigabbs.blogspot.com/
Re: dnetc benchmarks
« Reply #77 on: February 05, 2012, 01:31:31 PM »
Tl;dr
 

Offline number6

Re: dnetc benchmarks
« Reply #78 on: February 05, 2012, 02:03:16 PM »
Quote from: takemehomegrandma;679320

But the Generals leading the country (Hyperion) decides that this is not a "worthy enough" solution for Ethiopia(TM), that Ethiopia(TM) is all about being different and better (in some weird, backwards way), that what's needed is a "special Ethiopian(TM) dish", and only this will save the Ethiopian people.



I enjoyed both of your posts. Very creative and well written. However, your facts are incorrect.

In the case of Moana, as well as the earlier attempt in 2004 when not only a port was suggested, but a source of h/w had been presented along with the case for the port...
It was Amiga Inc. who decided against both propositions.

source

Does your story have enough room for the additional characters? Heh.

#6
 

Offline Zac67

  • Hero Member
  • *****
  • Join Date: Nov 2004
  • Posts: 2890
    • Show only replies by Zac67
Re: dnetc benchmarks
« Reply #79 on: February 05, 2012, 02:16:19 PM »
dnetc RC5-72 on AMD HD5750: 725,256,214 keys/s
ca. 100€ => 7+ MKeys/s per EUR
 

Offline Iggy

  • Hero Member
  • *****
  • Join Date: Aug 2009
  • Posts: 5348
    • Show only replies by Iggy
Re: dnetc benchmarks
« Reply #80 on: February 05, 2012, 02:54:35 PM »
Wow!
takemehomegrandma.

Now that's a rant.

As I've said before, I don't think Treavor's investment was a sound business decision.
That said, I don't see the harm in it and I like the basic design of the X1000.
Is it too expensive? Obviously, if you're designing a new, low volume PPC based system that's inevitable.
If I had the resources, I'd be tempted to do something similar (although I'd use one of Varisys' recommended processors from Freescale, not a dead end like the PA6T).

Having someone build a vanity system based on idealized specifications (that will probably never be a commercial success) is hardly the disaster you paint it out to be.
Will it greatly increase the numbers of people using AOS4?
Nope.

Would I take one if offered?

So fast your head would spin.
"Not making any hard and fast rules means that the moderators can use their good judgment in moderation, and we think the results speak for themselves." - Amiga.org, terms of service

"You, got to stem the evil tide, and keep it on the the inside" - Rogers Waters

"God was never on your side" - Lemmy

Amiga! "Our appeal has become more selective"
 

Offline itix

  • Hero Member
  • *****
  • Join Date: Oct 2002
  • Posts: 2380
    • Show only replies by itix
Re: dnetc benchmarks
« Reply #81 on: February 05, 2012, 03:09:41 PM »
Quote from: Tripitaka;679317
Of course, MOS will be faced with having to break out of only supporting old powerpc hardware at some point too. Those old Macs will only last for so long.


Lets put it this way: what Hyperion is going to do when PPC is finished?
My Amigas: A500, Mac Mini and PowerBook
 

Offline jorkany

  • Hero Member
  • *****
  • Join Date: Sep 2006
  • Posts: 1009
    • Show only replies by jorkany
    • http://www.amigaos4.com
Re: dnetc benchmarks
« Reply #82 on: February 05, 2012, 03:37:12 PM »
Quote from: TheDaddy;679259
@amigadave


Exactly. I totally I agree with you. This is an OS4 machine, for OS4 users is the DREAM machine, like the A4000 was for A500 users.


How so? I mean, if OS4 actually supported Xena, Xorro, the second CPU core, and the more common onboard hardware - yeah maybe it would be somebody's idea of a "dream". As it is though it's just a somewhat faster CPU & RAM.

Unless you mean, it's a dream to make OS4 look even more ridiculous, if so then point taken.
 

Offline jorkany

  • Hero Member
  • *****
  • Join Date: Sep 2006
  • Posts: 1009
    • Show only replies by jorkany
    • http://www.amigaos4.com
Re: dnetc benchmarks
« Reply #83 on: February 05, 2012, 03:39:55 PM »
Quote from: spirantho;679299
If AmigaOS 4 ever gets ported to the Mac, it will be doomed to stay on old hardware forever, as the market for new machines would be obliterated- which is exactly what they do not want.

Too late, OS4 is anchored onto PPC.
 

Offline minator

  • Hero Member
  • *****
  • Join Date: Jan 2003
  • Posts: 592
    • Show only replies by minator
    • http://www.blachford.info
Re: dnetc benchmarks
« Reply #84 on: February 05, 2012, 03:51:32 PM »
I posted a link to some benchmarks in this thread in post 228.

According to them the PA6T not only beats a G5 at a higher clock speed but it completely destroys the G4.

Looks like it is misconfigured to me.  Is the L2 switched on?  What power mode is it in?  Are both memory controllers on? etc...


BTW This is a remarkably common problem. Even the big boys do it.
 

Offline minator

  • Hero Member
  • *****
  • Join Date: Jan 2003
  • Posts: 592
    • Show only replies by minator
    • http://www.blachford.info
Re: dnetc benchmarks
« Reply #85 on: February 05, 2012, 03:55:05 PM »
Quote from: itix;679335
Lets put it this way: what Hyperion is going to do when PPC is finished?


Good question, and given all the PPC suppliers are switching to ARM, one they are going to have to answer fairly soon.
 

Offline PiruTopic starter

  • \' union select name,pwd--
  • Hero Member
  • *****
  • Join Date: Aug 2002
  • Posts: 6946
    • Show only replies by Piru
    • http://www.iki.fi/sintonen/
Re: dnetc benchmarks
« Reply #86 on: February 05, 2012, 04:19:06 PM »
Quote from: minator;679344
I posted a link to some benchmarks in this thread in post 228.

According to them the PA6T not only beats a G5 at a higher clock speed but it completely destroys the G4.

Looks like it is misconfigured to me.  Is the L2 switched on?  What power mode is it in?  Are both memory controllers on? etc...
Reading the actual slides you run into
Quote
Application included significant VMX (AltiVecTM) use and significant main-memory utilization
(emphasis mine)

PA6T has significant advantage when processing large data set from/to memory.

Some further comments:
  • 970FX is a single core CPU, 970MP would have crushed the PA6T, at least if it would have had enough memory bandwidth (but would naturally have been a power pig at that, but that's beside the point here). If the application really is as memory bound is it looks like, PA6T might have even won over 970MP.
  • 7447 of 975MHz was used. Considering the PLL multipliers available for the 7447 it means that the bus didn't even run at 166MHz, but likely something less. Considering the memory bound nature of the application it likely explains the extremely poor 7447 results. The later PowerStream G4 processor modules do run 154MHz MPX bus with 6.5 multiplier though, but even that is still extremely slow compared to 970 not to mention PA6T.
So I don't think it's up to misconfiguration, but rather CPU bound vs memory bound tasks. In Addition OS4 does not support the 2nd core of PA6T.
« Last Edit: February 05, 2012, 05:47:33 PM by Piru »
 

Offline Iggy

  • Hero Member
  • *****
  • Join Date: Aug 2009
  • Posts: 5348
    • Show only replies by Iggy
Re: dnetc benchmarks
« Reply #87 on: February 05, 2012, 04:43:03 PM »
Quote from: minator;679346
Good question, and given all the PPC suppliers are switching to ARM, one they are going to have to answer fairly soon.

Interesting opinion, but false.

Both AppliedMicro and Freescale continue to develop new PPC processors.

No current ARM processor outperforms even the dated G5.

Freescale has announce the re-introduction of AltiVec instructions to many of its PPC product lines and one processor that has as many as 20 virtual cores.

ARM is an in-order processor design with no where near the power of an out of order PPC design (per clock cycle a PPC is an significantly more powerful design).

All three of the most popular gaming platforms use PPC derived processors and at least two of their successors will continue to do so.
"Not making any hard and fast rules means that the moderators can use their good judgment in moderation, and we think the results speak for themselves." - Amiga.org, terms of service

"You, got to stem the evil tide, and keep it on the the inside" - Rogers Waters

"God was never on your side" - Lemmy

Amiga! "Our appeal has become more selective"
 

Offline zylesea

  • Hero Member
  • *****
  • Join Date: Feb 2006
  • Posts: 638
    • Show only replies by zylesea
    • http://www.via-altera.de
Re: dnetc benchmarks
« Reply #88 on: February 05, 2012, 06:16:35 PM »
Quote from: minator;679346
Good question, and given all the PPC suppliers are switching to ARM, one they are going to have to answer fairly soon.


Quite a bold statement about *all* ppc suppliers. Freescale isn't switching to ARM - they are supporting both, ppc & ARM since ages. And I don't see ibm going ARM in favor of ppc.
ARM has clearly most of the buzz and very probably a brighter future than ppc, but it's not a world revolution and ppc going >nil:

Offline commodorejohn

  • Hero Member
  • *****
  • Join Date: Mar 2010
  • Posts: 3165
    • Show only replies by commodorejohn
    • http://www.commodorejohn.com
Re: dnetc benchmarks
« Reply #89 from previous page: February 05, 2012, 06:19:17 PM »
Quote from: Iggy;679348
Interesting opinion, but false.

Both AppliedMicro and Freescale continue to develop new PPC processors.

No current ARM processor outperforms even the dated G5.

Freescale has announce the re-introduction of AltiVec instructions to many of its PPC product lines and one processor that has as many as 20 virtual cores.
But everybody says PPC is dead! If it's repeated so much, it must be true, right?
Computers: Amiga 1200, DEC VAXStation 4000/60, DEC MicroPDP-11/73
Synthesizers: Roland JX-10/MT-32/D-10, Oberheim Matrix-6, Yamaha DX7/FB-01, Korg MS-20 Mini, Ensoniq Mirage/SQ-80, Sequential Circuits Prophet-600, Hohner String Performer

"\'Legacy code\' often differs from its suggested alternative by actually working and scaling." - Bjarne Stroustrup