Welcome to MilkyWay@home

Separation Application Shutting Down on Tuesday, Jun 20th

Message boards : News : Separation Application Shutting Down on Tuesday, Jun 20th
Message board moderation

To post messages, you must log in.

Previous · 1 . . . 12 · 13 · 14 · 15 · 16 · Next

AuthorMessage
Mr P Hucker
Avatar

Send message
Joined: 5 Jul 11
Posts: 990
Credit: 376,143,149
RAC: 0
Message 76202 - Posted: 7 Jul 2023, 1:11:06 UTC - in response to Message 76197.  
Last modified: 7 Jul 2023, 1:12:16 UTC

Small cache? LMAO

Dude, Google is like... right here. www.google.com

3900X = 64MB L3 cache
3900XT = 64MB L3 cache
7742 (EPYC, 64-core, ROME) = 256MB L3 cache
And the dictionary is right there. I said "maybe". There has to be some reason it's rubbish at running the benchmark. Like I told Ian&Steve, a benchmark is just aprogram, and his CPU isn't very good at it.

As I have told you before and I'll say it again. Benchmarks are useless metrics. If you want to know how a CPU performs, then get real world data from what you actually intend to use the CPU for.
Not possible with so many Boinc projects.

The EPYC is good on every BOINC project. Not just Universe@home. I only picked Universe@home because I knew GPUUG users have EPYCs that run it nearly 24/7. So I knew I'd be getting good results from their EPYC hosts.
And yet it can't run benchmarks very well. There's something up with it.

You are right. EPYCs are good at one thing. Number crunching which is what 99% of these BOINC projects need. What they are NOT good at is gaming which is what 99% of all benchmarks revolve around.
Gaming is still numbers, angles of trajectories of light and moving objects. Just like er.... physics in LHC, Milkyway, Universe.
The above was double spaced between sentences, I apologise for the forum software ruining my post.
ID: 76202 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Skillz

Send message
Joined: 28 May 17
Posts: 76
Credit: 4,397,709,051
RAC: 337,066
Message 76203 - Posted: 7 Jul 2023, 1:32:21 UTC - in response to Message 76202.  

And the dictionary is right there. I said "maybe". There has to be some reason it's rubbish at running the benchmark. Like I told Ian&Steve, a benchmark is just aprogram, and his CPU isn't very good at it.


Do you run benchmarks all day? Is that your goal? Probably not. So who cares what benchmarks can do.

Keep reading everything. I already explained why benchmarks are not a good metric.

Unless you are gaming, doing video/image production, office suite programs, compression, or other niche things that most people do with their computers then a benchmark is useless.

I really don't know how else I can explain to you that benchmarks are worthless metrics for this stuff.

Not possible with so many Boinc projects.


Yes it is. We do it for all new hardware that comes out on my team. Someone buys the new shiny CPUs and we put it to the test on all the projects to see what they are capable of doing on those projects. Hint: 7950X is a beast of a CPU for a desktop, 16 core CPU. EPYC Genoa can't be beat. Although the EPYC Genoa CPUs are crazy expensive still being that those CPUs just came out recently. I think they'll set you back like $3k+ per CPU; but holy cow are they crazy good at BOINC.

And yet it can't run benchmarks very well. There's something up with it.


Ignorance is bliss.

Gaming is still numbers, angles of trajectories of light and moving objects. Just like er.... physics in LHC, Milkyway, Universe.


Sure, its math. But its a different type of math, different type of instruction set. Desktop/gaming CPUs are way better at gaming than server CPUs. It has almost always been this way. I don't know anyone with EPYC CPUs who are gaming on them though. Kind of dumb to buy an EPYC when there are far better gaming CPUs for a gaming rig. Same thing for office work.

You keep getting hung up on benchmarks. Yet I have showed you REAL WORLD data that shows the EPYC demolishing your Ryzen 9 3900X and 3900XT. Would you like me to find more BOINC projects that show the same thing? All you show me is some useless benchmarks.

Oh and let's not forget this part:


https://www.cpu-monkey.com/en/compare_cpu-amd_ryzen_9_3900xt-vs-amd_epyc_7742

I'd venture to guess that cpu-monkey doesn't even own an EPYC. Kind of hard to benchmark a CPU when you don't own one.
ID: 76203 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Profile Conan
Avatar

Send message
Joined: 2 Jan 08
Posts: 123
Credit: 69,556,860
RAC: 795
Message 76216 - Posted: 10 Jul 2023, 0:23:29 UTC

The thing with Bench marking, in relation to BOINC, as that is what we are doing, it serves less purpose than when originally designed.
With so many projects using fixed credit systems it is becoming redundant for use as a measure for granting credit.
However for the projects that still use them, bench marks are used for the granting of credits and in some cases with a quorum of two required then it is often the lower claim that is awarded.
This really comes to a head when you don't run bench marks as you get the default BOINC bench mark of 1 billion floating points and 1 billion drystones even if you have never run a bench mark in BOINC.

This will always give you a very low credit claim and your so called wingman will (even if they have run bench marks) get a low awarded credit score as the lowest is awarded.

So running a bench mark on the projects that use them does serve a purpose (for the granting of credit), for other BOINC projects that don't use them they serve no purpose and are not needed at all.

Bench marks are supposed to be part of the BOINC software for the use of the CreditNew system to award credit.
It does not work very well,
YAFU only uses the default unmodified BOINC software and CreditNew.
I recently ran a 16t work unit on my faster higher bench marking 7900X Ryzen system and it used all 24 threads for 9 hours (equal to about 7 days of processing) and got 1,450 odd points.
I then ran an 8t work unit on my lower bench marked Ryzen 5900X and it ran for 8 hours using up to 24 threads (equal to about 5 days processing) and it was awarded 54,200 odd points.

I can't figure that out, bench marks and CreditNew just don't work.

Bench marks are not the be all and end all for the testing of a CPU, as no single bench mark does every thing (at least as far as I have seen), it there was a single bench mark that could test every thing then they would not have to run a "Test Suite" to see how a CPU preforms they could just run one programme.

I don't need a bench mark to tell me that my Ryzen 5900X runs a programme at a slower pace than my Ryzen 7900X, it is obvious with the times taken.

My computers are not hidden you can see the Bench Mark difference but they don't tell the whole story.

Bench Marks, don't believe them. CreditNew, don't use it.

Rant over, thanks for your time.

Conan
ID: 76216 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Mr P Hucker
Avatar

Send message
Joined: 5 Jul 11
Posts: 990
Credit: 376,143,149
RAC: 0
Message 76217 - Posted: 10 Jul 2023, 1:14:14 UTC - in response to Message 76216.  

Bench marks are not the be all and end all for the testing of a CPU, as no single bench mark does every thing (at least as far as I have seen), it there was a single bench mark that could test every thing then they would not have to run a "Test Suite" to see how a CPU preforms they could just run one programme.
Why is a benchmark not simply running a test suite? When I find a list of benchmarks on a website, they should have run a variety of programs to come up with that value.
The above was double spaced between sentences, I apologise for the forum software ruining my post.
ID: 76217 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Profile mikey
Avatar

Send message
Joined: 8 May 09
Posts: 3328
Credit: 522,654,004
RAC: 85,804
Message 76218 - Posted: 10 Jul 2023, 3:05:39 UTC - in response to Message 76217.  

Bench marks are not the be all and end all for the testing of a CPU, as no single bench mark does every thing (at least as far as I have seen), it there was a single bench mark that could test every thing then they would not have to run a "Test Suite" to see how a CPU preforms they could just run one programme.


Why is a benchmark not simply running a test suite? When I find a list of benchmarks on a website, they should have run a variety of programs to come up with that value.


In the beginning because people didn't want to run something they weren't getting credit for, nor did it help the Science the Project was trying to do as everyone essentially ran the same mini task.
ID: 76218 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Mr P Hucker
Avatar

Send message
Joined: 5 Jul 11
Posts: 990
Credit: 376,143,149
RAC: 0
Message 76219 - Posted: 10 Jul 2023, 3:26:27 UTC - in response to Message 76218.  
Last modified: 10 Jul 2023, 3:26:42 UTC

Why is a benchmark not simply running a test suite? When I find a list of benchmarks on a website, they should have run a variety of programs to come up with that value.
In the beginning because people didn't want to run something they weren't getting credit for, nor did it help the Science the Project was trying to do as everyone essentially ran the same mini task.
I was asking why things like cpubenchmark.net don't use a program which tests all the aspects of a CPU.
The above was double spaced between sentences, I apologise for the forum software ruining my post.
ID: 76219 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Profile mikey
Avatar

Send message
Joined: 8 May 09
Posts: 3328
Credit: 522,654,004
RAC: 85,804
Message 76220 - Posted: 10 Jul 2023, 10:30:38 UTC - in response to Message 76219.  

Why is a benchmark not simply running a test suite? When I find a list of benchmarks on a website, they should have run a variety of programs to come up with that value.
In the beginning because people didn't want to run something they weren't getting credit for, nor did it help the Science the Project was trying to do as everyone essentially ran the same mini task.


I was asking why things like cpubenchmark.net don't use a program which tests all the aspects of a CPU.


Oh sorry, Toms Hardware does that
ID: 76220 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Mr P Hucker
Avatar

Send message
Joined: 5 Jul 11
Posts: 990
Credit: 376,143,149
RAC: 0
Message 76221 - Posted: 10 Jul 2023, 10:39:24 UTC - in response to Message 76220.  
Last modified: 10 Jul 2023, 10:39:39 UTC

Oh sorry, Toms Hardware does that
Does Tom's Hardware have benchmarks for the two aforementioned CPUs (mine and Ian&Steve's) which agrees with Boinc's speed?
The above was double spaced between sentences, I apologise for the forum software ruining my post.
ID: 76221 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Skillz

Send message
Joined: 28 May 17
Posts: 76
Credit: 4,397,709,051
RAC: 337,066
Message 76227 - Posted: 10 Jul 2023, 20:42:29 UTC - in response to Message 76217.  

Bench marks are not the be all and end all for the testing of a CPU, as no single bench mark does every thing (at least as far as I have seen), it there was a single bench mark that could test every thing then they would not have to run a "Test Suite" to see how a CPU preforms they could just run one programme.
Why is a benchmark not simply running a test suite? When I find a list of benchmarks on a website, they should have run a variety of programs to come up with that value.


No reviewers buy or get free samples of EPYC (or any server grade CPU) because that's not their target audience. The benchmark you think you found for the 7742, the site doesn't even own it. They're just "guessing" what it can do, which is the wrong way.

There is no such thing as a "end all, be all" software suite that can test every possible thing a CPU can do.

As I said before. Benchmarks you find on sites like that do not benchmark for the niche market BOINC is in.

They target gamers mostly, but also office work and audio/video production/editing.
ID: 76227 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Mr P Hucker
Avatar

Send message
Joined: 5 Jul 11
Posts: 990
Credit: 376,143,149
RAC: 0
Message 76230 - Posted: 10 Jul 2023, 23:27:22 UTC - in response to Message 76227.  

No reviewers buy or get free samples of EPYC (or any server grade CPU) because that's not their target audience. The benchmark you think you found for the 7742, the site doesn't even own it. They're just "guessing" what it can do, which is the wrong way.
I assumed if they haven't got their own, they use tests users have done with their software and average them. I know some do. I've run benchmark software on my own computer then it lets you compare with other people who have used it, so you know if your CPU is working properly.

There is no such thing as a "end all, be all" software suite that can test every possible thing a CPU can do.
Why on earth not? Just write a program which does each of the sorts of calculations proper programs do.

As I said before. Benchmarks you find on sites like that do not benchmark for the niche market BOINC is in.

They target gamers mostly, but also office work and audio/video production/editing.
They give a score for different types of calculation. Surely Boinc falls into one of those.
The above was double spaced between sentences, I apologise for the forum software ruining my post.
ID: 76230 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Skillz

Send message
Joined: 28 May 17
Posts: 76
Credit: 4,397,709,051
RAC: 337,066
Message 76231 - Posted: 10 Jul 2023, 23:59:51 UTC - in response to Message 76230.  

I assumed if they haven't got their own, they use tests users have done with their software and average them. I know some do. I've run benchmark software on my own computer then it lets you compare with other people who have used it, so you know if your CPU is working properly.


I don't know what your argument is. I posted a screenshot that explained how that web site got their benchmark totals. It claims to use some "top secret formula" which basically means, they have no idea. Just plugging in numbers. I avoid sites that don't go into detail on how they get their results because it's poppy cock at best.

Why on earth not? Just write a program which does each of the sorts of calculations proper programs do.


There are BILLIONS of possible combinations. You want to write something that can emulate all that. Go ahead.

Benchmark software developers are only going to write benchmarks for their target audience. As I have said, multiple times. They're just for gaming, office work and other similar stuff.

As I said before. Benchmarks you find on sites like that do not benchmark for the niche market BOINC is in.

They target gamers mostly, but also office work and audio/video production/editing.
They give a score for different types of calculation. Surely Boinc falls into one of those.[/quote]

I am sure some of the might cross. You have to have the CPU and run the benchmark first. No one has, to my knowledge. So good luck.
ID: 76231 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Mr P Hucker
Avatar

Send message
Joined: 5 Jul 11
Posts: 990
Credit: 376,143,149
RAC: 0
Message 76232 - Posted: 11 Jul 2023, 1:36:18 UTC - in response to Message 76231.  

I don't know what your argument is. I posted a screenshot that explained how that web site got their benchmark totals. It claims to use some "top secret formula" which basically means, they have no idea. Just plugging in numbers. I avoid sites that don't go into detail on how they get their results because it's poppy cock at best.
You said the sites were guessing, but they don't have to get a CPU themselves to test it, if the benchmark suite they use is available to the public, which a lot are, they can look at what scores users got.

There are BILLIONS of possible combinations. You want to write something that can emulate all that. Go ahead.
There are a manageable number of instructions types, eg AVX. There's cache size and speed. A reasonable number of things to look at.

Benchmark software developers are only going to write benchmarks for their target audience. As I have said, multiple times. They're just for gaming, office work and other similar stuff.
I doubt Boinc is all that different to other programs.
The above was double spaced between sentences, I apologise for the forum software ruining my post.
ID: 76232 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Profile mikey
Avatar

Send message
Joined: 8 May 09
Posts: 3328
Credit: 522,654,004
RAC: 85,804
Message 76235 - Posted: 11 Jul 2023, 11:23:20 UTC - in response to Message 76232.  

I don't know what your argument is. I posted a screenshot that explained how that web site got their benchmark totals. It claims to use some "top secret formula" which basically means, they have no idea. Just plugging in numbers. I avoid sites that don't go into detail on how they get their results because it's poppy cock at best.


You said the sites were guessing, but they don't have to get a CPU themselves to test it, if the benchmark suite they use is available to the public, which a lot are, they can look at what scores users got. .


The problem is do you run Boinc at 100% cpu usage or 90% or even 80 or less percent? That all needs to be factored in when you are using someone elses benchmark numbers. We can 'assume' alot of things but if you don't do it yourself there are ALOT of variables that can affect how benchmark numbers are arrived at.
ID: 76235 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Mr P Hucker
Avatar

Send message
Joined: 5 Jul 11
Posts: 990
Credit: 376,143,149
RAC: 0
Message 76239 - Posted: 11 Jul 2023, 12:43:11 UTC - in response to Message 76235.  

The problem is do you run Boinc at 100% cpu usage or 90% or even 80 or less percent? That all needs to be factored in when you are using someone elses benchmark numbers. We can 'assume' alot of things but if you don't do it yourself there are ALOT of variables that can affect how benchmark numbers are arrived at.
Those won't make that much difference. Anyway all I'm interested in is comparing CPU x with CPU y. The benchmark isn't like Boinc, it always runs 100%.
The above was double spaced between sentences, I apologise for the forum software ruining my post.
ID: 76239 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Profile Bill F
Avatar

Send message
Joined: 4 Jul 09
Posts: 87
Credit: 16,905,275
RAC: 4,045
Message 76240 - Posted: 11 Jul 2023, 18:14:18 UTC - in response to Message 76239.  

I you are interested in CPU-x verses CPU-y buy one of each and experiment to your hearts content. The rest of us mostly don't care.
ID: 76240 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Skillz

Send message
Joined: 28 May 17
Posts: 76
Credit: 4,397,709,051
RAC: 337,066
Message 76241 - Posted: 11 Jul 2023, 21:11:18 UTC - in response to Message 76235.  

I don't know what your argument is. I posted a screenshot that explained how that web site got their benchmark totals. It claims to use some "top secret formula" which basically means, they have no idea. Just plugging in numbers. I avoid sites that don't go into detail on how they get their results because it's poppy cock at best.


You said the sites were guessing, but they don't have to get a CPU themselves to test it, if the benchmark suite they use is available to the public, which a lot are, they can look at what scores users got. .


The problem is do you run Boinc at 100% cpu usage or 90% or even 80 or less percent? That all needs to be factored in when you are using someone elses benchmark numbers. We can 'assume' alot of things but if you don't do it yourself there are ALOT of variables that can affect how benchmark numbers are arrived at.


We're not talking about BOINC's built in benchmark. Which in itself is worthless. We're talking about benchmarking software such as Cinebench, 3DMark and others.

Some people are just too dumb to read about those benchmarks and see what they are actually benchmarking. They just look at the numbers like wow, that must be good.

Tired of repeating myself to someone who refuses to learn anything.
ID: 76241 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Mr P Hucker
Avatar

Send message
Joined: 5 Jul 11
Posts: 990
Credit: 376,143,149
RAC: 0
Message 76243 - Posted: 13 Jul 2023, 5:49:12 UTC - in response to Message 76240.  

I you are interested in CPU-x verses CPU-y buy one of each and experiment to your hearts content. The rest of us mostly don't care.
When they cost a grand each, that's a preposterous thing to do.
The above was double spaced between sentences, I apologise for the forum software ruining my post.
ID: 76243 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Mr P Hucker
Avatar

Send message
Joined: 5 Jul 11
Posts: 990
Credit: 376,143,149
RAC: 0
Message 76244 - Posted: 13 Jul 2023, 5:50:41 UTC - in response to Message 76241.  

We're not talking about BOINC's built in benchmark. Which in itself is worthless. We're talking about benchmarking software such as Cinebench, 3DMark and others.

Some people are just too dumb to read about those benchmarks and see what they are actually benchmarking. They just look at the numbers like wow, that must be good.

Tired of repeating myself to someone who refuses to learn anything.
And I'm tired of you not understanding it's very easy to write a benchmark which tests everything, then you could have a gaming score, a video editing score, a Boinc type score, etc, etc. All you need is a bunch of instructions of the type used in these projects.
The above was double spaced between sentences, I apologise for the forum software ruining my post.
ID: 76244 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Profile mikey
Avatar

Send message
Joined: 8 May 09
Posts: 3328
Credit: 522,654,004
RAC: 85,804
Message 76245 - Posted: 13 Jul 2023, 12:14:01 UTC - in response to Message 76244.  

We're not talking about BOINC's built in benchmark. Which in itself is worthless. We're talking about benchmarking software such as Cinebench, 3DMark and others.

Some people are just too dumb to read about those benchmarks and see what they are actually benchmarking. They just look at the numbers like wow, that must be good.

Tired of repeating myself to someone who refuses to learn anything.
And I'm tired of you not understanding it's very easy to write a benchmark which tests everything, then you could have a gaming score, a video editing score, a Boinc type score, etc, etc. All you need is a bunch of instructions of the type used in these projects.


But who would pay you to do the testing, companies don't care about Boinc scores, or this score or that score especially if it isn't similar to someplace elses scores, remember way back when when Intel started Celeron cpu's and the scores plummeted and AMD said 'our cpu's don't do that' and the market shifted to AMD cpu's. Then AMD saw the benefit to what Intel did and started making their scaled back versions of cpu's, they probably took them from the bin pile and retested them and sold them as Duron and Sempron cpu's but it was too late for Intel to grab the share of the market back that they had lost.
ID: 76245 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Mr P Hucker
Avatar

Send message
Joined: 5 Jul 11
Posts: 990
Credit: 376,143,149
RAC: 0
Message 76246 - Posted: 13 Jul 2023, 12:51:37 UTC - in response to Message 76245.  

But who would pay you to do the testing
You offer the program free to test home machines. Millions run it to check if their machine is running ok, the result is uploaded to the database, and they make a big list.

companies don't care about Boinc scores, or this score or that score especially if it isn't similar to someplace elses scores,
Well Epycs seem to have been designed for Boinc or something similar.
The above was double spaced between sentences, I apologise for the forum software ruining my post.
ID: 76246 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Previous · 1 . . . 12 · 13 · 14 · 15 · 16 · Next

Message boards : News : Separation Application Shutting Down on Tuesday, Jun 20th

©2024 Astroinformatics Group