The continuing of life's adventures of the kittyman.

Message boards : Cafe SETI : The continuing of life's adventures of the kittyman.
Message board moderation

To post messages, you must log in.

Previous · 1 . . . 28 · 29 · 30 · 31 · 32 · 33 · 34 . . . 52 · Next

AuthorMessage
kittyman Crowdfunding Project Donor*Special Project $75 donorSpecial Project $250 donor
Volunteer tester
Avatar

Send message
Joined: 9 Jul 00
Posts: 51478
Credit: 1,018,363,574
RAC: 1,004
United States
Message 1898097 - Posted: 29 Oct 2017, 17:42:54 UTC - in response to Message 1898095.  

Well, I shall let you know the results once it is installed and running.
M.2 bandwidth is surely a lot faster than SATA.
Asus claims 10Gb/second for SATA and 32Gb/second for M.2 transfer rates.
How much of an improvement in boot time that actually translates to remains to be seen.

Meow!
"Time is simply the mechanism that keeps everything from happening all at once."

ID: 1898097 · Report as offensive     Reply Quote
Profile Mike Special Project $75 donor
Volunteer tester
Avatar

Send message
Joined: 17 Feb 01
Posts: 34382
Credit: 79,922,639
RAC: 80
Germany
Message 1898098 - Posted: 29 Oct 2017, 17:43:33 UTC
Last modified: 29 Oct 2017, 17:47:44 UTC

It depends on the brand of M2 drive Keith.
My Samsung 960 EVO is significantly faster than the SSD drive.
3000 MB/S against 512 MB/S and booting into win 7 is another 10 seconds faster on my PC.
You need to make sure it runs at PCIe Gen 3x4.


With each crime and every kindness we birth our future.
ID: 1898098 · Report as offensive     Reply Quote
kittyman Crowdfunding Project Donor*Special Project $75 donorSpecial Project $250 donor
Volunteer tester
Avatar

Send message
Joined: 9 Jul 00
Posts: 51478
Credit: 1,018,363,574
RAC: 1,004
United States
Message 1898102 - Posted: 29 Oct 2017, 18:07:42 UTC - in response to Message 1898098.  

The M.2 connector on my board IS rated PCIE 3x4.
My 960 PRO M.2 card is rated up to 3.5Gb/s sequential read speed.
My 850 PRO SATA SSD is rated at up to .55Gb/s sequential read speed.

That is a lot of difference, depending on how the mobo deals with the capability.
"Time is simply the mechanism that keeps everything from happening all at once."

ID: 1898102 · Report as offensive     Reply Quote
Profile Keith Myers Special Project $250 donor
Volunteer tester
Avatar

Send message
Joined: 29 Apr 01
Posts: 13164
Credit: 1,160,866,277
RAC: 1,873
United States
Message 1898115 - Posted: 29 Oct 2017, 20:06:56 UTC - in response to Message 1898098.  
Last modified: 29 Oct 2017, 20:08:00 UTC

It depends on the brand of M2 drive Keith.
My Samsung 960 EVO is significantly faster than the SSD drive.
3000 MB/S against 512 MB/S and booting into win 7 is another 10 seconds faster on my PC.
You need to make sure it runs at PCIe Gen 3x4.

I changed from a Crucial M200 SSD drive to a Samsung 960 Evo M.2 drive in a PCIe Gen3X4 lane configuration and I barely noticed any difference in boot times. Maybe a second, nothing that my seat of my pants accelerometer noticed. I am not disputing the transfer rate. In the 1000's Mb/sec versus the 100's Mb/sec of the SSD. But again not much difference in perception of time to boot or times for programs to appear ready to use after clicking them. YMMV obviously.
Seti@Home classic workunits:20,676 CPU time:74,226 hours

A proud member of the OFA (Old Farts Association)
ID: 1898115 · Report as offensive     Reply Quote
kittyman Crowdfunding Project Donor*Special Project $75 donorSpecial Project $250 donor
Volunteer tester
Avatar

Send message
Joined: 9 Jul 00
Posts: 51478
Credit: 1,018,363,574
RAC: 1,004
United States
Message 1899308 - Posted: 5 Nov 2017, 15:28:54 UTC

Well.....
The new rig is finally pretty well settled in.
Went through a whole day of fighting with it to get my M.2 SSD properly installed and recognized.
But it is done and working well.
As was said, very little change in boot time. But the rig is now running on the M.2 and the old SSD is disconnected.

Added a lot of upgrade fans to the case. Much airflow. Those EVGA 140mm fans are the cat's meow!

The VRM temp is now stable at 45c.

The GPUs are at 52c (bottom card) and 63c (top card) at 2040Mhz full load.

Ram is at 3360Mhz with 14/14/14/28/42/1T timings.

CPU is at 4.41Ghz at 1.35vcore. Temp is from 49c to 54c depending on the load.

And it be crunching very well indeed.

Meeeeeeeeeeeeeeeeeeeeeeeowza!
"Time is simply the mechanism that keeps everything from happening all at once."

ID: 1899308 · Report as offensive     Reply Quote
kittyman Crowdfunding Project Donor*Special Project $75 donorSpecial Project $250 donor
Volunteer tester
Avatar

Send message
Joined: 9 Jul 00
Posts: 51478
Credit: 1,018,363,574
RAC: 1,004
United States
Message 1899317 - Posted: 5 Nov 2017, 15:42:41 UTC - in response to Message 1899316.  

215,000 RAC is not to be sniffed at :-))

Meow
Thank you, Chris.
The new rig is close to 89K by itself.
Don't think it will get much higher in it's current configuration.
"Time is simply the mechanism that keeps everything from happening all at once."

ID: 1899317 · Report as offensive     Reply Quote
Profile Keith Myers Special Project $250 donor
Volunteer tester
Avatar

Send message
Joined: 29 Apr 01
Posts: 13164
Credit: 1,160,866,277
RAC: 1,873
United States
Message 1899339 - Posted: 5 Nov 2017, 16:34:00 UTC

Damn, kittyman .... those are some nice tight timings for the memory. Sounds from the temps that you might have more room to push the cpu higher. Try for 4.6 or 4.7 Ghz?
Seti@Home classic workunits:20,676 CPU time:74,226 hours

A proud member of the OFA (Old Farts Association)
ID: 1899339 · Report as offensive     Reply Quote
kittyman Crowdfunding Project Donor*Special Project $75 donorSpecial Project $250 donor
Volunteer tester
Avatar

Send message
Joined: 9 Jul 00
Posts: 51478
Credit: 1,018,363,574
RAC: 1,004
United States
Message 1899342 - Posted: 5 Nov 2017, 16:39:28 UTC - in response to Message 1899339.  

Damn, kittyman .... those are some nice tight timings for the memory. Sounds from the temps that you might have more room to push the cpu higher. Try for 4.6 or 4.7 Ghz?

Dunno how high I dare push the vcore...........
"Time is simply the mechanism that keeps everything from happening all at once."

ID: 1899342 · Report as offensive     Reply Quote
Profile Keith Myers Special Project $250 donor
Volunteer tester
Avatar

Send message
Joined: 29 Apr 01
Posts: 13164
Credit: 1,160,866,277
RAC: 1,873
United States
Message 1899379 - Posted: 5 Nov 2017, 19:03:56 UTC - in response to Message 1899342.  

Did you just whack 1.35V in at the 4.45Ghz clock? Or did you methodically raise each parameter and test for stability. The 1.35V Vcore is in the nominal range I see reported for 4.6 Ghz in the forums. Can you just bump the multiplier to 45 and see if its stable at 1.35V? Then bump to 46 and test again. If there is free Mhz to be gained at no more thermal increase, I say go for it.

Or have you just got to the point you are done with the system and just said I'm going to use it where it's at. No harm in that either.
Seti@Home classic workunits:20,676 CPU time:74,226 hours

A proud member of the OFA (Old Farts Association)
ID: 1899379 · Report as offensive     Reply Quote
kittyman Crowdfunding Project Donor*Special Project $75 donorSpecial Project $250 donor
Volunteer tester
Avatar

Send message
Joined: 9 Jul 00
Posts: 51478
Credit: 1,018,363,574
RAC: 1,004
United States
Message 1899382 - Posted: 5 Nov 2017, 19:13:50 UTC - in response to Message 1899379.  

Did you just whack 1.35V in at the 4.45Ghz clock? Or did you methodically raise each parameter and test for stability. The 1.35V Vcore is in the nominal range I see reported for 4.6 Ghz in the forums. Can you just bump the multiplier to 45 and see if its stable at 1.35V? Then bump to 46 and test again. If there is free Mhz to be gained at no more thermal increase, I say go for it.

Or have you just got to the point you are done with the system and just said I'm going to use it where it's at. No harm in that either.

I'm actually at 126bclk x 35.................and 1.355vcore.
And I gradually worked up to that.
"Time is simply the mechanism that keeps everything from happening all at once."

ID: 1899382 · Report as offensive     Reply Quote
Profile Keith Myers Special Project $250 donor
Volunteer tester
Avatar

Send message
Joined: 29 Apr 01
Posts: 13164
Credit: 1,160,866,277
RAC: 1,873
United States
Message 1899410 - Posted: 5 Nov 2017, 22:00:14 UTC

OK, I keep forgetting you are using a BCLK generator. I have never had that option before so no experience with it. Any reason you went that route instead of keeping to default front side bus clocks and just using the multipliers for cpu and memory?
Seti@Home classic workunits:20,676 CPU time:74,226 hours

A proud member of the OFA (Old Farts Association)
ID: 1899410 · Report as offensive     Reply Quote
kittyman Crowdfunding Project Donor*Special Project $75 donorSpecial Project $250 donor
Volunteer tester
Avatar

Send message
Joined: 9 Jul 00
Posts: 51478
Credit: 1,018,363,574
RAC: 1,004
United States
Message 1899412 - Posted: 5 Nov 2017, 22:07:38 UTC - in response to Message 1899410.  

OK, I keep forgetting you are using a BCLK generator. I have never had that option before so no experience with it. Any reason you went that route instead of keeping to default front side bus clocks and just using the multipliers for cpu and memory?

Because it also overclocks other subsystems on the mobo, not just the CPU.
"Time is simply the mechanism that keeps everything from happening all at once."

ID: 1899412 · Report as offensive     Reply Quote
Profile Keith Myers Special Project $250 donor
Volunteer tester
Avatar

Send message
Joined: 29 Apr 01
Posts: 13164
Credit: 1,160,866,277
RAC: 1,873
United States
Message 1899435 - Posted: 6 Nov 2017, 0:23:09 UTC

I understand that. But you could just choose the 3200 or 3466 Mhz XMP profile for the memory and achieve the same throughput. And the danger of a high BCLK is that it also boosts the PCIe bus to the point that some PCIe cards and controllers flake out. I don't know if the PCIe bus for your motherboard can be locked to a bus speed independent of the BLCK generator. As I stated previously, I am a real newbie when it comes to Intel hardware. Thanks for the update.
Seti@Home classic workunits:20,676 CPU time:74,226 hours

A proud member of the OFA (Old Farts Association)
ID: 1899435 · Report as offensive     Reply Quote
kittyman Crowdfunding Project Donor*Special Project $75 donorSpecial Project $250 donor
Volunteer tester
Avatar

Send message
Joined: 9 Jul 00
Posts: 51478
Credit: 1,018,363,574
RAC: 1,004
United States
Message 1899461 - Posted: 6 Nov 2017, 3:59:21 UTC

Well......
About all I can tell ya is this.
After getting a new mobo after 8 years, I know less about this bios and chipset than most folks.
I likened the BCLK setting to the FSB (front side buss) setting that I did most of my OCing with on the old Intel chipsets.
So that is what I started playing with first.
This architecture is much more complex with some bits and pieces seemingly running on their own clocks independent of what I am doing with the CPU.
I have much to learn about it yet, but by trial and error I am learning some of what works and what simply crashes hard.
Many more adventures to come, I am sure.

Meow!
"Time is simply the mechanism that keeps everything from happening all at once."

ID: 1899461 · Report as offensive     Reply Quote
Profile Keith Myers Special Project $250 donor
Volunteer tester
Avatar

Send message
Joined: 29 Apr 01
Posts: 13164
Credit: 1,160,866,277
RAC: 1,873
United States
Message 1899472 - Posted: 6 Nov 2017, 5:48:43 UTC - in response to Message 1899461.  

But that is what is so fun with playing with new hardware. So many adventures!
Seti@Home classic workunits:20,676 CPU time:74,226 hours

A proud member of the OFA (Old Farts Association)
ID: 1899472 · Report as offensive     Reply Quote
kittyman Crowdfunding Project Donor*Special Project $75 donorSpecial Project $250 donor
Volunteer tester
Avatar

Send message
Joined: 9 Jul 00
Posts: 51478
Credit: 1,018,363,574
RAC: 1,004
United States
Message 1899532 - Posted: 6 Nov 2017, 14:33:00 UTC - in response to Message 1899475.  

Yeeeees spose so! :-) Still remember the days of the frozen Penny, now THAT was tinkering to the nth degree :-))

Indeed it was.
And I still have the chiller unit in the basement.
I'd love to crank it up and have a go with it again. But, I proved by going through a couple of CPUs due to condensation related corrosion that it was not viable for 24/7 running long term. I think I could get about 6 months or so at a time. Even with all my moisture control efforts by packing the CPU socket and the mobo front and back with liberal amounts of clear silicone grease, those little water molecules were bound and determined to get to that -30c cooled CPU. And they eventually did.
If and when I ever have the disposable income to do so, I would like to try it once more.
After installing the CPU and testing it, I would spray the mobo front and back around the CPU socket with some kind of coating that would seal it off more completely than the silicone grease did. I would mask the top of the CPU to keep it clear for installation of the chiller puck with heatsink compound. Of course, it would be a one-shot attempt, as you would never be able to get the CPU out of the socket again.
Maybe some day...................................

Meow!
"Time is simply the mechanism that keeps everything from happening all at once."

ID: 1899532 · Report as offensive     Reply Quote
Profile Gary Charpentier Crowdfunding Project Donor*Special Project $75 donorSpecial Project $250 donor
Volunteer tester
Avatar

Send message
Joined: 25 Dec 00
Posts: 31014
Credit: 53,134,872
RAC: 32
United States
Message 1899547 - Posted: 6 Nov 2017, 16:40:30 UTC - in response to Message 1899532.  

ID: 1899547 · Report as offensive     Reply Quote
kittyman Crowdfunding Project Donor*Special Project $75 donorSpecial Project $250 donor
Volunteer tester
Avatar

Send message
Joined: 9 Jul 00
Posts: 51478
Credit: 1,018,363,574
RAC: 1,004
United States
Message 1899555 - Posted: 6 Nov 2017, 17:13:35 UTC - in response to Message 1899547.  

Yeah, something like that. There are various types of conformal coatings made specifically for use on circuit boards. Some are mil-spec if I recall.
"Time is simply the mechanism that keeps everything from happening all at once."

ID: 1899555 · Report as offensive     Reply Quote
kittyman Crowdfunding Project Donor*Special Project $75 donorSpecial Project $250 donor
Volunteer tester
Avatar

Send message
Joined: 9 Jul 00
Posts: 51478
Credit: 1,018,363,574
RAC: 1,004
United States
Message 1900029 - Posted: 9 Nov 2017, 16:05:34 UTC
Last modified: 9 Nov 2017, 16:11:29 UTC

Hmmmm.
Now this is interesting. I have never heard of this before. But apparently it's been available since 2012.

Intel actually has a special warranty plan for overclockers. It is in addition to the standard 3 year warranty, and will replace a CPU one time that has been incapacitated due to overclocking.
The kitties said I should consider it.............
Might allow me to push a little harder than I would otherwise.
Intel's Performance Tuning Protection Plan.

Meowoverclockinginsurance.
"Time is simply the mechanism that keeps everything from happening all at once."

ID: 1900029 · Report as offensive     Reply Quote
Profile BMH
Avatar

Send message
Joined: 27 May 99
Posts: 419
Credit: 166,294,083
RAC: 125
United Kingdom
Message 1900039 - Posted: 9 Nov 2017, 16:43:38 UTC - in response to Message 1900029.  

Interesting indeed... but, are they able to determine if a CPU has been overclocked or not, should a failed one be returned for replacement? CPUs seem pretty resilient these days and in my experience it's the motherboard that suffers due to prolonged overclocks.

Incidentally, I've just bought a few new GPUs of the 1060 variety from Zotac; they have an extended warranty up to 5 years which swayed my decision, but I believe overclocking voids this warranty. This is a slight shame since a gentle overclock is unlikely to do much harm whereas poor airflow at standard speeds could certainly shorten the life. With GPUs I'm guessing that clock/memory speeds et al could be logged on the board and thus read by the manufacturer to get an idea how the component was set/performing prior to it failing.
Brian.
ID: 1900039 · Report as offensive     Reply Quote
Previous · 1 . . . 28 · 29 · 30 · 31 · 32 · 33 · 34 . . . 52 · Next

Message boards : Cafe SETI : The continuing of life's adventures of the kittyman.


 
©2024 University of California
 
SETI@home and Astropulse are funded by grants from the National Science Foundation, NASA, and donations from SETI@home volunteers. AstroPulse is funded in part by the NSF through grant AST-0307956.