SI
SI
discoversearch

   Technology StocksAMD, ARMH, INTC, NVDA


Previous 10 Next 10 
To: THE WATSONYOUTH who wrote (20837)10/3/2017 3:33:45 PM
From: neolib
of 21667
 
I have no idea. I wouldn't think embedded GPU announcements would move this much, I don't recall them doing so in the past. Its a very small market for AMD, the shift in market cap today would likely equal 5 years of embedded GPU sales, not to mention profits.

Share KeepReplyMark as Last Read


To: knykny who wrote (20841)10/3/2017 3:38:03 PM
From: neolib
of 21667
 
There were filings, they just didn't show what was claimed in that article. Those share counts were all stock options they exercised and sold immediately. Along with additional shares. Its the opposite of what that article claimed. There have been zero purchases on the open market of shares by execs at AMD. They have instead been unloading them.

Share KeepReplyMark as Last Read


From: Vattila10/3/2017 3:49:12 PM
of 21667
 

Interesting optimisation results for Ryzen (lower is better, left bars "Modified Vrad" show the improvement achieved by fixing a "false sharing" issue in Valve's Source engine; this issue leading to unnecessary inter-core communication, I presume):



forums.anandtech.com

Share KeepReplyMark as Last ReadRead Replies (1)


To: Vattila who wrote (20844)10/3/2017 4:01:35 PM
From: neolib
of 21667
 
Yipes, the original version slowed down with increasing thread count!

The "fixed" version still doesn't scale terribly well with thread count although at least the sign is right! Look at what doubling the count from 16 to 32 does.

Share KeepReplyMark as Last ReadRead Replies (1)


To: neolib who wrote (20845)10/3/2017 4:06:41 PM
From: Vattila
of 21667
 

The poster explains the poor scaling. It is probably due to Amdahl's Law, a good portion of sequential code that is not speeded up by parallelisation.

"The speedup in the fixed section of the code is likely a fair bit higher than what's shown here."

But, yes, the case shows how easy it is to introduce a scaling problem if the programmer does not understand multi-threading and shared memory well. Luckily, the shared counter, in this case, was useless and could be removed.

Share KeepReplyMark as Last ReadRead Replies (1)


To: THE WATSONYOUTH who wrote (20837)10/3/2017 6:26:57 PM
From: Pravin Kamdar
of 21667
 
trades almost dead flat for 4 days and then jumps 5+% in an hour.............we are just along for the ride...........question is..... who has got the wheel??
Don't know. I can't see from the end of the chain behind the car.

Sold on the way up today. Uncomfortable with only 33% of my full load shares.

Pravin.

Share KeepReplyMark as Last Read


To: neolib who wrote (20830)10/3/2017 6:31:15 PM
From: Pravin Kamdar
of 21667
 
Yeah, no reason to buy on the open market. They can, are, and will, just cut themselves as many zero basis options as they want -- for immediate sale.

Pravin.

Share KeepReplyMark as Last Read


To: Vattila who wrote (20846)10/3/2017 7:43:16 PM
From: THE WATSONYOUTH
of 21667
 
the case shows how easy it is to introduce a scaling problem if the programmer does not understand multi-threading and shared memory well.


......at what point does one no longer call the process "optimization" but an original "screw up" ??

Share KeepReplyMark as Last Read


To: bit3 who wrote (20839)10/3/2017 9:12:13 PM
From: THE WATSONYOUTH
of 21667
 
.........well.............that's how I intend to go................go to bed feeling fine......and wake up dead.........I think I'd prefer putting it off for a while though...............I'm 65

Share KeepReplyMark as Last ReadRead Replies (1)


To: THE WATSONYOUTH who wrote (20850)10/3/2017 10:00:29 PM
From: neolib
of 21667
 
Better make the most of the next year :)

Share KeepReplyMark as Last ReadRead Replies (1)
Previous 10 Next 10 

Copyright © 1995-2017 Knight Sac Media. All rights reserved.Stock quotes are delayed at least 15 minutes - See Terms of Use.