Originally posted by maldorordiscord
View Post
Announcement
Collapse
No announcement yet.
MIPS Loongson 3A Benchmarks On Debian
Collapse
X
-
Originally posted by TobiSGD View PostPlease show me the part in that text were the author states that he is translating SSE into Loongson's SIMD format. He is stating that he has done that that with MMX on a Loongson 2F, which was possible because it has a 64 bit SIMD unit that is somewhat similar to MMX. You still fail to give a proof for your claims. Also, I don't think to spend 19$ on an article that I think will also not contain a proof for your claims.
source and directlink to the article: http://www.google.de/url?sa=t&rct=j&...UkCyWNUjQXuoiA
the Loongson3A only get 70% of the speed of a native x86 core because it accelerate integer x86 and FPU-X87 and MMX and SSE1 and SSE2 in hardware.
Comment
-
Originally posted by maldorordiscord View PostI did your work the Loongson 3A can accelerate : integer x86 and FPU-X87 and MMX and SSE1 and SSE2 in hardware.
source and directlink to the article: LINK EDITED OUT TO PREVENT DAMAGE TO THE FORUM DUE TO COPYRIGHT INFRINGEMENT
the Loongson3A only get 70% of the speed of a native x86 core because it accelerate integer x86 and FPU-X87 and MMX and SSE1 and SSE2 in hardware.
Sadly, the shown graphs say nothing about the SSE performance since there is no information about compiler options and which SIMD units (or the FPU) were used during the benchmarks.
I would also like to see a complete benchmark comparison with an mainstream x86-CPU, like the AMD A8 or a Phenom II X4 to see how fast those CPUs are in reality and if gaming would be an option, which I still doubt.Last edited by TobiSGD; 08 July 2012, 07:04 PM.
Comment
-
Originally posted by TobiSGD View PostSo finally you came up with a proof for at least a part of your statements and it took you only 3 days to do so.
And I'm sure these haters in this forum will ban me for showing you the truth.
Originally posted by TobiSGD View PostSadly, the shown graphs say nothing about the SSE performance since there is no information about compiler options and which SIMD units (or the FPU) were used during the benchmarks.
I would also like to see a complete benchmark comparison with an mainstream x86-CPU, like the AMD A8 or a Phenom II X4 to see how fast those CPUs are in reality and if gaming would be an option, which I still doubt.
But one is for sure the Loongson is the most advance MIPS CPU on planet earth and in my point of view its also more advance than all ARM chips together.
Comment
-
Originally posted by maldorordiscord View PostThe source is the same you just don't read it because you don't want to pay 19 dollar. It took me 3 days to find a "pirate copy" edition to save you 19 dollar.
And I'm sure these haters in this forum will ban me for showing you the truth.
Comment
-
Originally posted by TobiSGD View PostRather strange that you have to find a pirated version of an article you already have. Also nice that you mentioned that this is a pirated version of that article, so that I could edit out the link from the quote I made to prevent damage to the forum due to copyright infringement.
And no I don't had the article before I found the free one in the web.
I just interpolated the information with logic based on a minimal basic information set.
My argumentation was based on interpolating informations with logic.
I told you already that I'm very good at thinking things complex.
And interpolating informations in a unknown area is complex thinking
Comment
-
Originally posted by maldorordiscord View PostLOL its funny you blame me for not showing you the source and after you consume the source for free because of my research about the article in the internet you blame me for copyright infringement.
Nice, Sherlock, very nice.
And no I don't had the article before I found the free one in the web.
I just interpolated the information with logic based on a minimal basic information set.
My argumentation was based on interpolating informations with logic.
I told you already that I'm very good at thinking things complex.
And interpolating informations in a unknown area is complex thinking
Comment
-
Originally posted by TobiSGD View PostYou do really think that this is funny? Yes, I blame you for copyright infringement, you posted a direct link to a PDF without stating that this is a pirated version of the copyrighted article you mentioned in an earlier post. That is exactly what copyright infringement means. With that you can damage Phoronix and every one that actually downloads that PDF.
Nice, Sherlock, very nice.
Originally posted by TobiSGD View PostYeah, exactly, you can guess what is written in an technical article before reading it and then use it as proof for your facts without even knowing the content of the article. This is complex lying, not complex thinking.
O man you are so simple in "Mind"
Compared to your simple-linear-thinking I'm really a High-Complex-Ingenious-genius.
You are pathetic limited.
Compared to you I'm a real Master of Mind.
Comment
-
Originally posted by maldorordiscord View PostI'm so sorry you deserve it.
LOL "complex lying" but the original article cover my "complex lying" this turns it into Intellectual giftedness.
O man you are so simple in "Mind"
Compared to your simple-linear-thinking I'm really a High-Complex-Ingenious-genius.
You are pathetic limited.
Compared to you I'm a real Master of Mind.
1. The Loongson 3 can emulate SSE in hardware.
2. It can handle 8 SSE instructions at a time.
3. It does this with 70% of native speed.
You have only proven number 1 to be true with this article. 33%. Mastermind? With having only 1 of 3 statements right? You are a master of overestimation of your own capabilities, but not a master of mind. You can't even do this simple math, but I am simple in mind, absolutely.
If one here is pathetic, it is definitely not me.
Comment
-
Originally posted by TobiSGD View PostYou don't even know me, but think I would deserve it to be accused for copyright infringement for your stupid actions?
You worked hard for it to teach me who you are and what is your need.
Originally posted by TobiSGD View PostAnd what is with the damage you may do to Phoronix with such actions? Does Michael deserve it also?
Originally posted by TobiSGD View PostYou are a moron and I hope you will soon get banned, if not for being Qaridarium circumventing the moderator decisions then for such dumb actions.
Originally posted by TobiSGD View PostOnly a third of your claims were proven to be right by that article. You have made 3 statements:
1. The Loongson 3 can emulate SSE in hardware.
2. It can handle 8 SSE instructions at a time.
3. It does this with 70% of native speed.
You have only proven number 1 to be true with this article. 33%. Mastermind? With having only 1 of 3 statements right? You are a master of overestimation of your own capabilities, but not a master of mind. You can't even do this simple math, but I am simple in mind, absolutely.
If one here is pathetic, it is definitely not me.
This means the 3A can handle 2 SSE(64bit) instructions at the same time per core.
but the 3A do have 4 cores 4*2=8
But the 3B do have 2*256bit simd-vector units per core and the 3C do have 2*512bit vector-simd units.
the 3A can already handle 8 SSE instructions at a time with all cores together.
the 3C can handle 256 ---> 64bit-sse instructions at a time.....
but you will still claim that the loongson 3C can not handle 8 pieces of 64bit-SSE instructions at the same time...
This means I'm 66% right. And the last one "It does this with 70% of native speed." has no relevance right now because to validate this we need real hardware and more real benchmarks.
Comment
Comment