• A Great friend to the HardForum with a great kid that he is trying to get a scholorship to continue his schooling. Please give hime a vote! Only 24 hours left! Thanks.
    If you have an VOTE FOR KEENAN!

GPU3 = Yawn | SMP = w00t

Status
Not open for further replies.
S

SpeedyVV

Supreme [H]ardness
Joined
Sep 14, 2007
Messages
4,210
Man, GPU3 has become reaaaaally boring with only 6800s. :(

Mind you SMP has been nothing but bigadvs left and right. :D
 
What rig do you have pulling bigadv's? I was under the impression that anything less than a dual socket board couldn't swing them. I wonder if my sig rig could push one out....
 
R-Type said:
What rig do you have pulling bigadv's? I was under the impression that anything less than a dual socket board couldn't swing them. I wonder if my sig rig could push one out....
Click to expand...
The only architecture with one processor that can manage completing all -bigadv WUs would be an i7 clocked ~3.6Ghz or more. These WUs require a minimum of 8 threads just to receive them.
 
R-Type said:
What rig do you have pulling bigadv's? I was under the impression that anything less than a dual socket board couldn't swing them. I wonder if my sig rig could push one out....
Click to expand...

Speedy is using a VM to run bigadv on his X6...I had no idea we were creating a monster when we concocted the idea..:D

 
ccityinstaller said:
Speedy is using a VM to run bigadv on his X6
Click to expand...

Uhh being that we're helping start this advisory board and are working on enforcing the rules............I would assume that our team having members circumventing the rules would be frowned upon? :confused:
I was under the impression that backending the client to run bigadv on anything less than 8 cores was against the rules?
Not trying to start a war here or be a boy scout; I'm just bringing up a point.
 
zero2dash said:
Uhh being that we're helping start this advisory board and are working on enforcing the rules............I would assume that our team having members circumventing the rules would be frowned upon? :confused:
I was under the impression that backending the client to run bigadv on anything less than 8 cores was against the rules?
Not trying to start a war here or be a boy scout; I'm just bringing up a point.
Click to expand...

Yes, running bigadv on a X6 is not allowed by the new guidelines as it doesn't have 8 threads and it require a hack to receive them.
 
Xilikon said:
Yes, running bigadv on a X6 is not allowed by the new guidelines as it doesn't have 8 threads and it require a hack to receive them.
Click to expand...
+1

I was under the impression that Speedy had stopped doing that though.....
 
Xilikon said:
Yes, running bigadv on a X6 is not allowed by the new guidelines as it doesn't have 8 threads and it require a hack to receive them.
Click to expand...

Unless those rules were changed just recently, Stanford allowed it per the VM client, but not the modded Windows native client:confused::confused:
 
Just out of question. What is he using as his VM for running it? Just a VM'd windows install, or an app like notfred? Just wondering as I'm running X5570's using notfred's linux appliance on ESXi 4.0, and would like to be able to enable bigadv if possible. (4 blades, each dual X5570's, which should have the appropriate horsepower to run them).
 
Syran said:
Just out of question. What is he using as his VM for running it? Just a VM'd windows install, or an app like notfred? Just wondering as I'm running X5570's using notfred's linux appliance on ESXi 4.0, and would like to be able to enable bigadv if possible. (4 blades, each dual X5570's, which should have the appropriate horsepower to run them).
Click to expand...

he has a simple VMWare image that was created with a file that allows 6 cores to download big adv.

edit: i was under the impression that technically by the rules it says 8 cores not threads, so the only "legit big adv folders" would be those with dual quad/hex core machines.
 
Syran said:
Just out of question. What is he using as his VM for running it? Just a VM'd windows install, or an app like notfred? Just wondering as I'm running X5570's using notfred's linux appliance on ESXi 4.0, and would like to be able to enable bigadv if possible. (4 blades, each dual X5570's, which should have the appropriate horsepower to run them).
Click to expand...


back when we spent 40+ hours straight working on it. it was VM+ubuntu+wine+ window smp client. it has a lot of overhead on it though. only reason we did this was to trick the client into thinking there was 8 threads.

none of this is required in your case though. all you need is a VM(one that supports 8+ threads) and run the windows client or run wine and the windows client.
 
This is one of the gray areas that I hope Pande clarifies for us with the guidelines.
 
Syran said:
Just wondering as I'm running X5570's using notfred's linux appliance on ESXi 4.0, and would like to be able to enable bigadv if possible. (4 blades, each dual X5570's, which should have the appropriate horsepower to run them).
Click to expand...
I don't know how it's done specifically but -bigadv isn't supported with the Linux SMP client natively nor is it supported in a Linux VM with a Windows host (it was originally). In your case, you will need to run a VM with Windows would be my guess per sirmonkey's suggestion.

W.Feather said:
edit: i was under the impression that technically by the rules it says 8 cores not threads, so the only "legit big adv folders" would be those with dual quad/hex core machines.
Click to expand...
That's a minor technicality and currently still permissible.
 
Wow, didn't realize there was such a debate on this issue...
 
Syran said:
Just out of question. What is he using as his VM for running it? Just a VM'd windows install, or an app like notfred? Just wondering as I'm running X5570's using notfred's linux appliance on ESXi 4.0, and would like to be able to enable bigadv if possible. (4 blades, each dual X5570's, which should have the appropriate horsepower to run them).
Click to expand...

It is an Ubuntu server VM running Wine and the Windows 6.30 client. Your machines don't need the Linux + Wine part, so a native Windows VM would be fine. The problem is that you will only be able to use 8 cores/threads unless ESXi supports more than that. 8 threads of a stock dual X5570 machine may not be enough to complete a bigadv unit in time. I haven't messed with type of setup to know for sure, though.
 
^^ What musky said

As long as i7 Quads and Hexes are running bigadvs I have no issue running X6 bigadv.

/* puts on flame retardant suit.
 
Xilikon said:
Dr Kasson already made a official statement about that : http://foldingforum.org/viewtopic.php?f=55&t=15882&p=157380#p157380
Click to expand...

Is this a mistake or are you intentionally misleading folks? :mad:

That is an old statement. Go to the last post on that thread for a further clarification.

Let me quote for those too lazy to click ;)

A clarification has been requested here. The purpose of bigadv is to run specific systems that are often large and slow and where we typically want a few relatively long trajectories rather than many shorter ones. As a result, the number of work units is often (though not always) limited, and we want to incentivize the fastest turnaround possible.

Bigadv was really designed for systems with >=8 cores. Manipulating the system to run bigadv on systems with 6 physical cores is possible; we would discourage it but aren't at the moment actively declaring it cheating if done without modifying the client. HOWEVER, we really want to emphasize the design for systems with >=8 cores. Just because someone can make the deadlines on current bigadv projects with certain 6-core systems is no guarantee that it will be possible in the future. If you're trying to game the system, complaining when the manipulation stops yielding you results is bad form. We'd really rather encourage more people to run bigadv with 12, 24, 48, etc. cores.

As mentioned above, there are a lot of more sophisticated things we could do for allocating machines and WU's. We'd love to be able to do them; some have technical issues, and in general we're trying to run a robust system with limited programming resources. And our primary goal is to do good science, with the help of our donors. We have some resources dedicated to infrastructure, but a lot of other work involves taking people away from science to work on infrastructure.
Click to expand...

Dont worry, I will NOT complain when the manupulation stops yielding results.

And if you are doing bigadvs with less than 12 physical cores, you dont need to stop either. But dont complain if they change the system.

See, I was right. GPU3 was where the fun was a while back because it was giving good points and some people where like you dont have fermi so stop....now the fun is with bigadvs and some people are like you dont have intel so stop....

I know this is not about AMD vs Intel.... but it is fun...

Dont forget, at the end of the day, the WU is still being returned on the timeframes they are looking for, and running an old crappy dual amd\intel box would have technically 8 plus physical cores, but would be slower and really defeating the results they are looking for.

SMP w00t.
 
Last edited:
SpeedyVV said:
As long as i7 Quads and Hexes are running bigadvs I have no issue running X6 bigadv.
Click to expand...

For all intents and purposes any os along with the client sees a 4 or 6 with HT as 8 or 12 cores. It doesn't matter if they're real silicon or just split threaded.

A hexa with no HT is still a hexa.

SpeedyVV said:
And if you are doing bigadvs with less than 12 physical cores, you dont need to stop either.
Click to expand...

There's no reason to stop. An HT i7 is still seen as 8 cores and that is the requirement. Until there is a migrant shift in os architecture and the os itself differentiates between threads and cores (which will never happen), HT i7s will be acceptable.

PG needs to quit dilly dallying and put some concrete rules in place to cover stuff like this instead of being wishy washy and pulling the "we don't like it but we're not going to outright condone it either" card. They just want the work turned in because the ends justify the means (apparently).

SpeedyVV said:
now the fun is with bigadvs and some people are like you dont have intel so stop....
Click to expand...

Are you really going to try to turn this into an Intel vs AMD discussion as a grounds for why it's ok to skate the rules? Come on man. :rolleyes:
 
Which part of 6 physical cores dont you understand?

Besides, what PG group needs to do is give results oriented rules. Who cares how many physical, virtual, whatever cores.

I can build a crap 12 core box and it would technically be good enough.

What if Intel comes out with a totally diff architecture that performs 100 times faster than an i7 Hex but technically has 1 physical core. Who gives a whoot.
 
I understand what 6 physical cores are. I own 2 of them, I think I know what that means. :rolleyes:

The fact is you're circumventing the 8 core check of the client by running a VM and tricking the client. You didn't hack the client, you just hacked the system running the client.
 
zero2dash said:
I understand what 6 physical cores are. I own 2 of them, I think I know what that means. :rolleyes:

The fact is you're circumventing the 8 core check of the client by running a VM and tricking the client. You didn't hack the client, you just hacked the system running the client.
Click to expand...
which apparently isn't forbidden by Kasson
 
jebo_4jc said:
which apparently isn't forbidden by Kasson
Click to expand...

Which is exactly why PG needs to quit playing pancake and put out some decisions and stick with them. FFS the thread in question is titled Is Folding -bigadv on less than 8 threads cheating? [Yes]

I mean is it just me or is the hypocrisy making this place rank?
Our team and others want to put a board together, we want to point out some things and make changes.... well it's hard to believe any effort will ever be made because in this case we have people on our team admittedly cheating the system (in addition to people still running GSO's with Fermi WUs because they want a boost).....so is something going to get done about any of this, or should we all just look the other way and stare at the floor and whistle while we walk away?? :mad: My point is ~ someone needs to make a damn decision on some of this stuff because skirting rules is not something I'm very tolerant of. :eek:
 
zero2dash said:
Which is exactly why PG needs to quit playing pancake and put out some decisions and stick with them. FFS the thread in question is titled Is Folding -bigadv on less than 8 threads cheating? [Yes]

I mean is it just me or is the hypocrisy making this place rank?
Our team and others want to put a board together, we want to point out some things and make changes.... well it's hard to believe any effort will ever be made because in this case we have people on our team admittedly cheating the system (in addition to people still running GSO's with Fermi WUs because they want a boost).....so is something going to get done about any of this, or should we all just look the other way and stare at the floor and whistle while we walk away?? :mad:
Click to expand...
Hang on, who is still running Fermi WU's on g92? If there is somebody out there we need to have that discussion again.

If you are just throwing around baseless accusations you should watch yourself because lurkers will come over here and take your ramblings for truth and use them against our whole team. We deserve better than that.

it appears the "[Yes]" was added by Bruce. Judging by Kasson's statement however, Bruce is incorrect.

The correct answer to put in the brackets would be "[No, but it is discouraged]"

Unless Kasson et al want to come out and say "Do not run bigadv on Phenom X6" I don't see anything wrong with what Speedy is doing. Kasson specifically left the door open.

It seems on different than what is the current policy with bigadv and i7. It is recommended/designed to run on 8 physical cores, so they don't recommend running it on i7, but we all have overclocked i7s that run bigadv just fine.
 
oh zero2dash, I am sorry you think that if the rules are not what you think they should be, it is hypocrisy.

Funny, when I was the one that said Pande Group rules are the ones that matter!

Get over it! What YOU think PG needs to do and what PG does are two different things.

And FFS you can start a thread titled Is my penis 12" long? [Yes]. But it does not make it true. BTW, i dont want to find out either. ;)
 
jebo_4jc said:
Hang on, who is still running Fermi WU's on g92? If there is somebody out there we need to have that discussion again.

If you are just throwing around baseless accusations you should watch yourself because lurkers will come over here and take your ramblings for truth and use them against our whole team. We deserve better than that.
Click to expand...

[sigh]
Soooooo not trying to be the whistleblower. :(
http://hardforum.com/showpost.php?p=1036376774&postcount=32
http://hardforum.com/showpost.php?p=1036377627&postcount=35

SpeedyVV said:
oh zero2dash, I am sorry you think that if the rules are not what you think they should be, it is hypocrisy.
Click to expand...

You apparently don't even know what the word means so I'll explain it more simply -
1) members of the [H]orde want to put forth a set of rules
2) a board is appointed across several teams
3) meanwhile members of the [H]orde are circumventing the rules
4) I see no mention in the "board topics" thread regarding shutting down the cheaters and/or putting a clear cut clarification forth of what is and is not acceptable

If our team wants to put up rules then we need to enforce them. I thought the consensus of our team was that we did not want people running G92 cards to run -advmethods and DL Fermi WUs, and likewise I was also under the impression that our team was not allowing anyone without 8 threads+ to run bigadv for our team. If those 2 statements are true (and I believe that they are), therein lies the hypocrisy. [H] cheaters are putting a black eye on our entire team and I don't like being labeled because other people want to skirt the rules or outright cheat when they're not supposed to.

We want to bitch about EVGA "cheating" by having a rewards program yet we're not going to keep our own team members from cheating themselves. :rolleyes:
I'm not above the team, and I'm not trying to be...I'm just calling it like I see it. If I'm wrong and there's no unofficial [H] rule about bigadv on X6, then by all means point it out and I'll gladly say I was incorrect and I'll also feel better about it. I thought the whole process in of itself started and finished as an experiment and not a plan.
 
Last edited:
You are jumping to the conclusion that since nothing has been said re: sirmonkey's running Fermi WU on his GS that the team is condoning his actions.

Personally I was not aware that this was going on, and I can assure you the team will deal with it to the extent we can. If you have been paying attention to the leadership of this team, both unofficial and official, you would know the team won't stand for that behavior. As such, I suggest you keep your hypocrite accusations quiet until you see a true response.
 
jebo_4jc said:
You are jumping to the conclusion that since nothing has been said re: sirmonkey's running Fermi WU on his GS that the team is condoning his actions.
Click to expand...

Maybe it was unknown, who knows. I don't know.
I'm bringing it up because I saw that. Not trying to get him or anyone in trouble; likewise I'm not trying to start a bunch of crap with anyone from the team either. I guess I'm playing KWH's role in this thread. :D -> We want to have rules, great, let's have them. Let's enforce what we can when we find out what we do, that's all I'm saying. I'm not intentionally trying to jump to conclusions.

jebo_4jc said:
you would know the team won't stand for that behavior.
Click to expand...

I know Zero82z would have definitely said something but apparently he's banned now? :confused: :eek:
 
regarding the X6/bigadv issue, I was under the same impression you were until I read Kasson's comment today.

The only "rule" we have in FAH is the rule that Kasson, Pande, et al provide. As such, if Kasson says something is permissible, that means it is permissible.

If we do have an unofficial "team rule" against running bigadv on X6, it would appear that rule needs to be changed in light of Kasson's comment
 
zero2dash, there are so many things wrong with your post that I am just going to ignore it and let you be happy.

But there is 1 thing I must point out as a downright LIE!

The EVGA Team are NOT cheaters. You do NOT represent the [H]orde and until you apologize to EVGA for implying they are cheaters, i wish you would just go away.

And dont give me the crap that "We want to bitch about EVGA "cheating"". I might not like it that a corporation sponsors a community based competition, but there are absolutely no rules against it, so please apologize, before you misrepresent the [H]orde by impliying they are cheaters.
 
Ok, so I started a thread to have some fun discussion on GPU vs SMP.

I am sorry, I am trying to have fun while folding. Unfortunately this seems like an impossebility lately within the DC forum.

I will close this thread. You guys want to argue about rules and non rules please do it somewhere else. This thread is as good as dead.
 
I apologize. I closed the thread before allowing zero2dash to reply.

Please go ahead zero2dash. I will close it after.
 
I just wanted to say ~ I'm not calling EVGA cheaters. Yeah I was pissed off when they overtook us but everyone was. Some people on our forums were throwing around the word "cheaters" because of their rewards program, just like BWM and a few others were bitching about our team having help from other teams (as payback for donations during the chimp challenge).

Thanks for reopening this thread so I could post that.

I apologize if I offended anyone with any of my comments.....not trying to, I love being a part of our team and I love all of you no matter what you run or don't run. /done ;)
 
Status
Not open for further replies.
Back
Top