Big Adv Drought?

[jebo_4jc]

Actually, I think that's exactly what might be going on with this so-called shortage, assuming that task isn't completely automated. It might not be what some are saying it is and work to generate new WUs just hasn't been done in a timely manner.

Right. In the past vj and kasson indicated bigadv shortages were caused by a deficiency in the labor necessary to inject Wu into the queue.

 

[APOLLO]

Right. In the past vj and kasson indicated bigadv shortages were caused by a deficiency in the labor necessary to inject Wu into the queue.

I see... Well, it's exactly what I was suspecting was the problem even though I didn't know of Vijay's past statements on the matter. There's really nothing we can do on our end to ameliorate it except hope they can get things better organized this year.

 

[86 5.0L]

pulled a 6900, YES!

 

[jebo_4jc]

I see... Well, it's exactly what I was suspecting was the problem even though I didn't know of Vijay's past statements on the matter. There's really nothing we can do on our end to ameliorate it except hope they can get things better organized this year.

Agreed. Of course, there's no guarantee that is the source of the current shortage, but I have my suspicions.

 

[Wiseguy2001]

It's pouring over here. Complete bigadv, a minute later... collect another biggy.

To be honest, it's not ideal. I was planning to fit in a bit of gaming this weekend

 

[capreppy]

I've been seeing Big Adv pretty consistently now. Just dropped a P2692 on SR2 and picked up P6900. I'll drop another Big Adv in 5 hours on my secondary i7.

There doesn't seem to be much of a shortage now, but if and when there comes a time that there is a shortage, I'll crunch A3 SMP without issue. I really do wish that the P670x are reevaluated though as they are a total PITA much like the P2684 Big Adv WUs.

 

[may i be worthy]

Right. In the past vj and kasson indicated bigadv shortages were caused by a deficiency in the labor necessary to inject Wu into the queue.

Well, it looks like some scientists got back from their holidays and filled the queue alright. Been back folding 48 hours and have got 3 consecutive batches of 3 x 6900s.

When you can't even pull a bigadv project older than 2 months, let alone an SMP, I think the drought is well and truly over. More of a blip really, in hindsight. But wouldn't it have been lovely to see a scientist blog along the lines of "we are working on it, more expected in a week or two?"

 

[Wiseguy2001]

A blog would be an excellent idea, not for just status updates but news of the direction F@H is going and what is in the is in the pipeline.

 

[capreppy]

Definitely a "blip". Been running Big Adv since my last posting on this thread.

 

[Patonb]

Its got to have been the 2600ks booting up and sucking up the WUs.

At 50k ppd who wouldn't!

 

[Activate: AMD]

Well, it looks like some scientists got back from their holidays and filled the queue alright. Been back folding 48 hours and have got 3 consecutive batches of 3 x 6900s.

When you can't even pull a bigadv project older than 2 months, let alone an SMP, I think the drought is well and truly over. More of a blip really, in hindsight. But wouldn't it have been lovely to see a scientist blog along the lines of "we are working on it, more expected in a week or two?"

Oh how quaint... somebody who thinks scientists have enough free time to write a blog (i promise from experience they don't)

probably just some overworked grad-students forgot to flip the switch on the bigadv machine

 

[Patonb]

Okay I call Shannaigianz!!

I've been a grad student... Engineer to boot...... They are not overworked.

Grad school was great!

 

[amdgamer]

Its got to have been the 2600ks booting up and sucking up the WUs.

At 50k ppd who wouldn't!

I think this right here was the biggest reason for the shortage.With the release of the 2600K's, it probably caused a spike in the number of folding machines that are now capable of running bigadv. As more and more machines become bigadv capable, I sometimes wonder whether we are going to start seeing a backlog of regular A3 work units build up like we have seen with the uniprocessor classic work units.

Oh how quaint... somebody who thinks scientists have enough free time to write a blog (i promise from experience they don't)

probably just some overworked grad-students forgot to flip the switch on the bigadv machine

You don't have to write a blog as it can be something as simple as a Facebook status update or a Tweets.

 

[jebo_4jc]

I've received two 2684s in the past 2 days. So stop hogging all the good WUs.

 

[Patonb]

I always wonder if its possibled to lump lots of uni-core into a bigger unit and release them.

I run 1 uni-core client on my laptop and i net like 160ppd on it.... if i leave it on.

But I always wonder how many uni-core mavchines are really out there.

 

[APOLLO]

Its got to have been the 2600ks booting up and sucking up the WUs.

At 50k ppd who wouldn't!

There were shortages in the past before SB processors were released. What do you attribute the cause to been then? It was stated by project managers last year that the 'shortage' is a labor problem rather than a supply problem to use industry terms. If there's a shortage of any kind it was on the manpower side not materials. I don't believe we as consumers/customers suddenly added to that in the past month.

I think this right here was the biggest reason for the shortage.With the release of the 2600K's, it probably caused a spike in the number of folding machines that are now capable of running bigadv...

I could be wrong but seriously doubt it. I don't see a mad dash to gobble up SB from the shelves and out of all the SB systems sold thus far, how many were realistically added to this DC project? Now let's compare that number to all the machines that have been running -bigadv since 2H '09. There are probably still more SR-2 systems alone than SB on this team I'd wager. Again, I could be wrong but I just don't see it unless all the SB guys are keeping eerily quiet. /shrug

 

[Patonb]

I'm sure if one or 2 people per team fired up a 2600k to run bigadvc it'd put an unexpected draw.

Stanford probably didn't expect the influx, so they didn't hasve the units ready.

Most resalistically its a combo of both

You forget that this team is an annomilly. Theres more sr-2s here than anywhere else. Your minimum top 20 is more than most full teams. So a few new 2600k here means very little, but over all its alot.

 

[APOLLO]

I'm sure if one or 2 people per team fired up a 2600k to run bigadvc it'd put an unexpected draw.

Stanford probably didn't expect the influx, so they didn't hasve the units ready.

Most resalistically its a combo of both

You forget that this team is an annomilly. Theres more sr-2s here than anywhere else. Your minimum top 20 is more than most full teams. So a few new 2600k here means very little, but over all its alot.

We don't have at our disposal raw purchase numbers for SB systems added to every team. Only then can we draw a more accurate conclusion that a specific architecture is the cause of the 'drought.' Without those numbers it's pure speculation. Yet, despite this we have an admission from a project founder that the contrary is true, and a labor shortage was responsible in the past. I would lean towards accepting his statement, that's all I'm saying.

 

[Activate: AMD]

Okay I call Shannaigianz!!

I've been a grad student... Engineer to boot...... They are not overworked.

Grad school was great!

Must have been lucky, because in all the labs I or my friends have been in, its a pretty much constant stream of work. I think engineering might be different than a phd student in hard science (i'm a chemist)..

Either way, glad to see the bigadv drought has passed.. though I was sad to see that my first unit after getting regular SMP's was a 2684

 

[may i be worthy]

Oh how quaint... somebody who thinks scientists have enough free time to write a blog (i promise from experience they don't)

probably just some overworked grad-students forgot to flip the switch on the bigadv machine

I am not asking for daily posts that take hours to write. I am talking about essentially a tweet. Anyone, repeat anyone can make that sort of time say twice a month. You can tweet during otherwise downtime like commuting or on the potty. (tweet while you toot?) We have the technology.

I've received two 2684s in the past 2 days. So stop hogging all the good WUs.

Oh alright. I have grabbed one large 2684 with my happy meal of two 6900s. Better?

I'm sure if one or 2 people per team fired up a 2600k to run bigadvc it'd put an unexpected draw.

Stanford probably didn't expect the influx, so they didn't hasve the units ready.

Most resalistically its a combo of both

You forget that this team is an annomilly. Theres more sr-2s here than anywhere else. Your minimum top 20 is more than most full teams. So a few new 2600k here means very little, but over all its alot.

The other benefit of being an anomaly (hey I like the idea of being one of those ) is that we also have people here that do more bigadv than just about anyone. When you see 4 new bigadv a day, and keep score, you can see patterns. The drought began with 6900s (~58% of units for the last 10 weeks) drying up. (well really just going from approx 2/3s of units to 1/3 of bigadv) 2692s were quadrupled, but it was not enough, and we saw SMPs a lot.

I was off doing other things for a week, but on my return I have had eleven 6900s and one 2684 in 3 days. Somebody turned the 6900 tap back on.

My guess is that if we were seeing some massive spike in production due to Sandy Bridges the ratio of work units to each other would not change dramatically. (just see more SMPs) But ratios of projects that had been very stable for 8 weeks did change, and right after the silly season. I think this is 90% about supply of units, and not demand. Like Apollo, I don't see enough chatter about SB here or on EVGA forums to think there would be a measurable spike, not compared to the massive effects that the 6900 project has had.

 

[Activate: AMD]

I am not asking for daily posts that take hours to write. I am talking about essentially a tweet. Anyone, repeat anyone can make that sort of time say twice a month. You can tweet during otherwise downtime like commuting or on the potty. (tweet while you toot?) We have the technology.

Just making a joke sheesh!

I agree, it'd be nice to get a few more details about whats going on under the hood, but as long as I keep getting my flow of 6900's then I can't complain too much

 

[APOLLO]

The other benefit of being an anomaly (hey I like the idea of being one of those ) is that we also have people here that do more bigadv than just about anyone. When you see 4 new bigadv a day, and keep score, you can see patterns. The drought began with 6900s (~58% of units for the last 10 weeks) drying up. (well really just going from approx 2/3s of units to 1/3 of bigadv) 2692s were quadrupled, but it was not enough, and we saw SMPs a lot.

I was off doing other things for a week, but on my return I have had eleven 6900s and one 2684 in 3 days. Somebody turned the 6900 tap back on.

My guess is that if we were seeing some massive spike in production due to Sandy Bridges the ratio of work units to each other would not change dramatically. (just see more SMPs) But ratios of projects that had been very stable for 8 weeks did change, and right after the silly season. I think this is 90% about supply of units, and not demand. Like Apollo, I don't see enough chatter about SB here or on EVGA forums to think there would be a measurable spike, not compared to the massive effects that the 6900 project has had.

I didn't look at the ratio of the various -bigadv WUs but you are 100% right. The mere fact that the download ratio dramatically changed between P6900s vs all other WUs lends additional credence to the theory that creation (or lack thereof) of new WUs was more the culprit than anything else. Although I will cede that a combination of factors could have possibly contributed. I guess we might never truly know. In any event it seems over for now, thankfully.

 

[jebo_4jc]

I guess we might never truly know.

the unfortunate reality of FAH.

IMO, better communication should be one of the primary goals of the DAB.

Ironically, I have heard very little about the DAB since its inception.

le sigh