Where are all the Core17 units?

[HaloJones]

See title! My Nvidia cards are getting nothing but 7xxx Core15 units no matter what client-type setting is used. Not only does this use more power for less points it introduces instability. Yeah, I know I could reduce the clocks. Yeah, I know there is no guarantee of particular units. But Vijay himself said to raise the issue if there is no units so here I am, raising the issue.

My production is halved and on the supposed basis that points=science, I'm doing less science for more cost.

[7im]

This is an exception to that rule of points equals science. The points difference is procedural not the assumption of slow folding. Core 15 is not on the QRB points system. Core 17 is. As you already know, the same thing happened back when CPU work units transitioned from the previous point system to the QRB system.

And not getting core 17 does not indicate a lack of those WUs. It indicates core 15 is set at a higher priority on the servers. To know if there is a lack of core 17 WUs, you would need to check the WU count on that work server.

[HaloJones]

And how does one tell which server is the Core17 work server?

[ChristianVirtual]

http://fah-web.stanford.edu/psummary.html
Look for project 8900, the current only core17 available ... 171.64.65.69

[Mstenholm]

he he. There are no information about avail WUs for that server so please check facts before ask others to do it.

[ChristianVirtual]

Saw that too, wondering too. Seems not updated in that stats ... Some helping mod might can jump in ...

[bollix47]

I do not have any numbers for WU availability for that server ... I see what you see. There is a % Assigned column that appears to be fairly active. Also, it was reported in another thread that more work units were created very recently ... IIRC it happened late Friday afternoon PST.

If you're 'stuck' on a core 15 server try setting the GPU slot to Finish. When it's done and uploaded Pause any other slots, quit and restart the client.

[PantherX]

Regarding the availability of FahCore_17 Projects, here is some information from proteneer over at the IRC:

Code: Select all

--- Day changed Wed Nov 27 2013
...
18:30 < rhavern> any ETA or "soon"(tm)?
18:31 <@proteneer_> haha, no :P
18:31 <@proteneer_> we're not slacking off :P don't worry
18:31 < rhavern> fair enough.  anything else you can talk about in the pipeline?
18:41 <@proteneer_> no :P
18:41 <@proteneer_> something
18:41 <@proteneer_> will come
18:41 <@proteneer_> in around January
18:41 <@proteneer_> early February
...

Of course, this may change if the situation is no longer the same. The point is, work is being done and will be released when ready. I do hope that it gets done sooner than later but considering that it is the Holiday Season, researchers and developers, just like the rest of us, do need a break so a new year for new projects would be a realistic expectation.

Regarding the Server 171.64.65.69, am not sure why the WUs avail isn't being shown. However, if you look at the % Ass G column, you will realize that of the last 500 WUs being assigned, 202 WUs were FahCore_17. Moreover, their priority is 10,000 (high) which is the same as non-FahCore_17 WUs Server (171.64.65.105 and 171.67.108.142). Just note that the last 500 GPU WUs include all currently supported GPUs (FahCore_11, FahCore_15, FahCore_16, FahCore_17).

[7im]

he he. There are no information about avail WUs for that server so please check facts before ask others to do it.

The lack of data for that one server does NOT change the standard process of how to find the WUs available. I gave good info. Not my problem if Standford screwed up. That's their problem. Criticize them for not following standard procedures.

[DocJonz]

Strangely, the GPU's I have running under Linux have continued with Core 17 WU's, while the ones under Windows have all switch to Core 15 WU's over the last few days - is that an AS side-effect?

[HaloJones]

Linux can only do Core17 so no surprise there. Maybe all the 17's are being given to the Linux community. For now I've shut down.

[7im]

Strangely, the GPU's I have running under Linux have continued with Core 17 WU's, while the ones under Windows have all switch to Core 15 WU's over the last few days - is that an AS side-effect?

Not a side affect. It's on purpose. As HJ said, Linux can only run core 17s. If the core 17s run low, the Linux GPUs fold nothing. At least the Windows systems can keep folding if core 17s run low.

[DocJonz]

I'd got so used to Linux using Core 15 etc. under WINE with v6 that I forgot the v7 client can only take Core 17 - Doh!

[HaloJones]

Whoever's in charge of this, you need to understand what is happening! Look across the forums of the major teams! on top of the disaster that is the announcement over BA, others are complaining at the loss of Core17 units. Cards are being shutdown and points contributions are falling for most of the top teams. It's no good saying "Oh, it's Christmas, there'll be more in January or February" People will have moved on and won't come back. FAH is down to <250K folders now. How many more can be lost?

[7im]

I guess those top teams won't be tops any more if they can't explain to their members the simple concept about never being guaranteed a steady supply of any type of WU. Or that changes to BA were announced 2 years ago, and should have been expected. Last weeks BA announcement is a just a continuation of both the original announcement, and the reminder that happened last year.

Where does the disappointment come from when they are pre-warned about these things?