Thanks for the info Steve,
I guess I have to look for someone in Norway that builds such things.
(there would be huge shipping costs and probably issues with the customs as well)
Looking forward to your results :)
Printable View
Thanks for the info Steve,
I guess I have to look for someone in Norway that builds such things.
(there would be huge shipping costs and probably issues with the customs as well)
Looking forward to your results :)
:up: Time to see some 5GHz+ stable runs.Quote:
Originally Posted by SteveRo
Once you have your cascade - what - 5.5-6 linx stable on 32nm?Quote:
Originally Posted by lowfat
SteveRo and Mike, i'm a bit curious here, what is your PCmark Vantage HDD score? And could you post SS of the HDD sub-test?
Also, Mike, what storage setup are you using now?
I am hoping I can run F@H @ 5GHz 24/7. But this is w/ a i7 920. Since VMWare only supports 8 threads there is no use for me to buy a 980x.Quote:
Originally Posted by SteveRo
this is nuts!Quote:
I am hoping I can run F@H @ 5GHz 24/7.
do u think it would manage to last even a year..?
I think so. I have a damn good chip. It ran @ 4.8GHz w/ only 1.3375V (24/7 full load of course). This was with a very poor single stage.Quote:
Originally Posted by onex
I think I should be able to do 5GHz and still keep it under 1.45V.
nice challenge going on there fellas :up:
Good morning,Quote:
Originally Posted by GullLars
Mike posted a score here, not sure if this is his fastest though.
Huge HDD score - 138077 - http://service.futuremark.com/compare?pcmv=302906
My best - trying to beat lowfat's ioxtreme and losing - 74868 - http://service.futuremark.com/compare?pcmv=285729
All the subtest scores for both are available at the above locations.
But Mike is not saying what storage he is using :D Classic :)Quote:
Originally Posted by SteveRo
You are still above him in my books. 138k is very obviously non legit/cached/whatever (non sustainable), especially considering it can not be ran on dynamic raid.Quote:
Originally Posted by SteveRo
The 1GB++/s bandwidth (if legit) means the only possibility is an LSI card and none of them are capable of that score.
pure conjecture here - but what about a bootable texas systems or intel pcie direct connect prototype or an areca 1880 4GB preproduction. Mike - can you say?Quote:
Originally Posted by One_Hertz
Hmmm... I've seen 74.000 in HDD score from 3R0 x25-V off ICH10R with 16KB stripe...
In our Norwegian SSD benchmark thread, we have the following PCmark Vantage HDD score listings above 100K points:
Anvil with 120.596 from 4R0 X25-M G1 off ICH10R.
Anvil with 120.374, 3R0 X25-M G1 off ICH10R.
Nizzen with 112.441 from 7R0 OCZ Vertex off Areca 1680ix.
Ourasi with 110.935 from 2R0 x25-E off ICH10R.
Anvil with 104.231 from 2R0 X25-M G2 160GB off ICH10R.
Perhaps 3-4 x25-E off ICH10R could give 130K+?
We should hope Areca makes their 18xx series with a more powerfull controller than LSI put in their 92xx series, since LSI 92xx can "only" do 80-90K IOPS on integrated RAID-0.
yes but were those results on bootable arrays? or with each device configured as individual arrays via ich10r? those scores look very much like ddr3 cache numbers....
Bootable arrays as in standard RAID-0 using the ICH in my case, but not running as boot drives.
yes so they wouldnt be able to run a real vantage benchmark then if they were configured as smaller arrays that were non-boot.
Anvil, the question is, what stripe size did you use for those arrays, and was WriteBack cache on?
Also, have you tried putting 8 (or all your) Intel SSDs (80/160GB) on your LSI 9260 and running PCmark vantage on that? Could be interresting, now that you can get 80-90K IOPS after new firmware.
Now, that is a good question, it's all hidden in the Norwegian thread :)
I haven't tried running Vantage on anything but ICH, I might give it a try though.
I'm a bit p****d at OCZ at the moment, they've announced that the Vertex 2 Pro wil be released shortly.
(so much for their words on the LE being exclusive)
gullars the 8R0 vertex on 9260 is tenth place vantage,
and 8R0 c300 is fourth (mbreslin)
would be interesting to see 8R0 intels!
the remarks i was making involving ddr3 cache refers to a trick used around pcm05 quite a bit to leverage the amount of system ram (ddr3) that is given to arrays on the ICH10r. it will produce uber results akin to those mentioned. it is just a matter of getting more assigned to a particular arrray. but that should not be able to be ran on a full vantage run, which is why you wont see the HDD test numbers equate to performance in the real benchmark. with numbers like those above^^^those guys should all be in the pcmark hall of fame. easily.
i have bigger results with 8R0 9211 than SteveRo. im using 0Mb of cache and SteveRo use 4Gigs and im still faster than him... Im using 8 cores instead of 6 but im slower..
The thing with pcmark Vantage is how do u setup ur pc. Using a 5Ghz cpu may not be faster than a 4.2-4.4Ghz .. Pc's are like cars maybe Mike has found his golden peak point that works for him. And yeah each pc handles different so maybe this is way his not speaking about his setup. In past records SteveRo used 12R0 acards but mike used only 3 ... Everyone here has a record but those records are not with more than 5 disk.. have u ever thought that? A pc is a pc there is nomatter how fast it can go.. the operator makes the laps not the autopilot!
+1 to mikes record not for being no1 but for show me that records doesnt match with bugets ;)
Yes, Tilt, concur with all you said - the 1231 cache makes up for it's other shortcomings (not being pcie 2.0, limited throughput, ...)
The PCM vantage gods seem to like areca controllers and it seems to be more than the controller cache.
its not about ARECA's or LSI's.. Even with that where is the C300 performance?Quote:
Originally Posted by SteveRo
Why do i kill C300 for fun even without cache..????? i score more with my drives than with c300's?? Anvil scored more even with 4.... You r looking in the wrong place to get back to top1 the 1rst place is not where the PCIE v2 or v3 is....
But how the pc works..how it responce... how to setit up currectly.
It doesnt matter what ur driving but it matters how u r driving it ;)
Now according to ur post PCIE has nothing to do with PCmarkV
and how do i know that??? just look the ICHxR performance... Limited to 750M/s top.. now with the areca even u r using 1 drive even 12 u will get the same results.. ALL READ IN CACHE :clap: :clap:
DDR3 cache ramdrive numbers are 250k+ I know because I tested :)Quote:
Originally Posted by Computurd
yes, ramdrive numbers are. but that is not the way it is being used , but i am sure you are aware of that. it is NOT being used as a ramdrive.
Anvil maybe you could give some quick breakdown of any settings/tweaks you do in w7 to get the most out of your array? tilt now and then spits out a tweak or 2 but I don't have the time to check a million things 1 by 1. (yes I know settings for intels won't necessarily work great for my c300s but it will show me I'm on the right track)Quote:
Originally Posted by Anvil
Mike giving a rundown of his would be amazing clearly he knows how to tweak!
Either of you not wanting to share your secrets is understandable too :) Just thought I would ask.
Good morning,
Having a bit more success with pcmark vantage, moving up a bit - http://service.futuremark.com/compare?pcmv=305488 :):):)
I think I will need better cooling to catch Mike -
http://img710.imageshack.us/img710/3061/clb.png
http://img534.imageshack.us/img534/1...katnumber2.png
Great achievement Steve,
I browsed the results yesterday, what brought you the last 300 points, fine-tuning or OC?
Thanks!Quote:
Originally Posted by Anvil
Fine-tuning and a bit of luck that I wouldn't fail the cpu gaming test (again).
I'm on the hairy edge of instability.
wow that is totally awesome man. great job just moving right on up! great work!:clap:
considering that corsair labs ran it at 5.8 you beating them at 4.8 is just outstanding
Congratz. Fun to see the top5 going:
1. Mike
2. Steve
3. *Company* labs.
4. Nizzen 24/7 (EDIT: and from Aug 09)
5. mbreslin
I'm hoping nizzen will pass corsair when he gets a new CPU and Areca 1880, possibly with ATI 6xxx if it's after the summer.
BTW, steve and mike, are your scores doable at 24/7 settings? If not, how high are your scores with your 24/7 settings?
I'm also wondering, are any of you guys in top5 sponsored? And have you asked for sponsors?
I hope Nizzen blows all his money on women and beer and can't upgrade. I should be ready to start throwing scores up by the weekend. My 980 isn't even installed yet.
Edit: I had no budget and I wasn't sponsored. Disposable income ftw.
Nizzen got a woman, and a kid too i belive...
BTW, have you tried a 6R0 on ICH10R? It could be interresting. It seems like PCmark Vantage likes 16KB stripes, possibly 8KB too, we're investigating a bit on the norwegian forum.
The pro of using ICH10R is that you aren't IOPS limited, untill you hit like 400MB/s (or higher). We have listed Anvil with 117K 4KB random read IOPS @ QD 64, with 1 worker (4R0 x25-M). That's like 468MB/s... He's also listed at 103K IOPS with 3R0 x25-M.
I just keep thinking of the IOPS from 6R0 C300 on ICH10R... You will max its bandwidth at 4KB random :P You could do 300K IOPS with 6 C300, but that would be 1200MB/s bandwidth...
To be honest I've never even hooked up more than 1 drive on ich, I went right to 9211 and then on to 9260, sometimes using 1 on ich for os and 7 in r0 on the lsis. I can try it, the problem becomes trying to run cabling in my case, even temporarily it's not easy. I have 10 5.25 bay devices and 10 drives, 2 360 rads and 1 480, 5970/5870/sound card/nic/9260, literally every single slot is filled. Throw in all the tubing and "permanent" type cables and even a mm ascension runs out of space.
wow, steve, that is superb!
on 4.8, man, U'R almost up there!!! again!!
just a little bit more, imagine what u can do with 5.5 and up with this array!!
u can beat them all :lol2: :up:!!
beautiful!!
great accomplishment:)!
p.s - lowfat, great spirit! way to go on this 5Ghz
mbreslin, what about just letting the 6 drives lay loose on the bench with the side-door off, or if you need to close the door, let them lay on the bottom of the cabinet or on top of other HDDs. There is no point in mounting them if you're not going to move the case.
You can also go McGuyver and use some duct tape against the door of the case, or any other surface.
When you hook it up, plz make a new thread, i have some suggestions and a couple of tests i'd like you to run. (among other things a IOmeter setup wich should take only a couple of minutes)
If you also have the possibility later of running OS from a single drive on 9211/9260, it could be cool to do stripe sizes 4 through 256KB and look how it effects PCmark Vantage HDD scores.
Thanks.Quote:
Originally Posted by Computurd
Also regarding 5.8 vs. 4.8 - I think Planet didn't use 4GB cache and wasn't allowed to use acards.
I can do 4.5 with vcore and qpi/vtt under 1.4v (intel max) and with temps?Quote:
Originally Posted by GullLars
Temps probably under 60C at full load - pcmv score for this is approx 31K.
I have no sponsorship - the closest I have to sponsorship is above average tolerance in my wife.
Thanks.Quote:
Originally Posted by onex
Once lowfat gets his cascade I expect him to post some new huge oc's into the 4.5 linx stable club
He has a top performer 920 chip that is waiting.
i don't understand u steve, are u going to use his system to cool u'r chip?
lol, i've got a 920 week 49 waiting here boxed for a mobo and some RAM sticks ;).
I think you are referring to Lowfat's cascade - I had previously suggested I could go to Lowfat's - that in jest.Quote:
Originally Posted by onex
If Lowfat was closer that would be a great idea.
I wonder how much trouble I could get into trying to handcarry a bag full of acards onto an airplane bound for Canada.
I do have a single stage phase change on the way from sdumper - hopefully have it in a couple weeks.
lol, with U.S ground check and laws, probably a lot :lol2:.Quote:
I wonder how much trouble I could get into trying to handcarry a bag full of acards onto an airplane bound for Canada.
@GullLars: I mount my ssds because the computer is supposed to be 24/7, hopefully getting to the point where I don't need to open the case anymore :p. I'm rebuilding loops this week for 980x, I will connect 6 extra sata cables to ich, shouldn't be too hard to switch them even with hard drives mounted. I did run 9260 with 1 drive on different stripe sizes, but I didn't ever go smaller than 64k. Also I didn't specifically run pcmv hdd suite as you're asking, I just ran the 3 iometer tests anvil linked me. 64k was the clear winner but maybe smaller will do better in vantage, I'd be willing to try.
For the 30GB Kingston see - http://www.bit-tech.net/hardware/201...128gb-review/1
"Results Analysis
With Toshiba having had 18 months on the SSD sidelines watching the competition learn and optimise their controllers, we were hoping for a solid showing. For the most part, we weren’t disappointed.
Sequential speeds as recorded in AS SSD averaged at 222MB/sec read and 177MB/sec write, and showed no deterioration following our 1TB write load and subsequent TRIM command. The sequential write performance was very good in particular, outpacing the Indilinx-powered OCZ Vertex by a commendable 30MB/sec.
Access times weren’t as unaffected by our 1TB write, rising from 0.139ms to 0.212ms for read and from 0.392ms to 0.439ms for write. However, these times are still pleasingly low, and comparable to competing SSDs.
Sequential performance measured in ATTO Disk Benchmark was also good, and while the new Toshiba controller shares the Indilinx drive controller’s poor 4KB sequential read performance, 64KB and 1,024KB performance was much better, with the 64KB sequential reads comfortably outpacing the Samsung controller in the Corsair P128.
One cause for concern with the Kingston is the sequential write performance with 64KB and 1,024KB files – we saw some minor, but noticeable, performance degradation following our writing of 1TB to the drive, despite the successful execution of the TRIM command. This is an issue we’ve also observed with the Samsung drive controller. While this issue won’t reduce performance by a horrendous amount, it still isn’t the perfect TRIM-based performance recovery displayed by the Intel and Indilinx drives.
What’s perhaps most surprising though is the comparatively poor random read and random write performance. We saw a random read speed of 18.8MB/sec, which is just half that of the OCZ Vertex and similar Indilinx-controlled drives. While a random write speed of 6.79MB/sec when clean and of 4.59MB/sec post-TRIM is a little better than the Samsung controller used in the original SSDNow V+, maximum write latencies, are worryingly high from the outset at 590ms, and peaked at 768ms post-TRIM. This is close to a whole second of random write lag, and approaches the sort of maximum write latencies we thought had been eliminated since the first generation of JMicron drive controllers.
Finally, looking to our remaining real world test of booting Windows 7, the drive performed identically in clean and TRIMed state, with a boot time of 34 seconds. This is roughly the same as all the competing high end drives, with the extra second of waiting likely due to the comparatively inferior read speeds of the Kingston.
@mbreslin-
your case will never stay closed, your on the bleeding edge bro :up:
If I were in ks or you were in cali I'd make you come over and do it for me ct ;p
SteveRo, any significant difference between using 2Gb cache and 4Gb?
and i would do it mr.breslin !
Seems like only in pcmv but in pcmv it was a big improvement.Quote:
Originally Posted by Jor3lBR
Seems like the suite score went up 1-2K.
Probably dependant on drives used and the speed of the rest of the machine.
I think a large cache can really improve an even bad array.
Not that 2x x25-v R0 is bad but it is nothing compared to 12xacard9010R0, yet the pcmv scores are only 10k apart or so (pcmv HDD test) when using 1231ml4g.
stevero
im expecting to see similar numbers from u!!!! i know what is ur new toy ;)
LSI HBA 9211 R0 with 8xIntels G2
http://img89.imageshack.us/img89/8678/bench1k.jpg
Tilt - wow, nice numbers - all are excellent except i would of expected seq write closer to 8*80=640 - closest is AS SSD of 530?
What stripe size was best for the x25m's and can you now adjust the stripe size?
Seems like I couldn't adjust the stripe size when i had the 9211 last December.
u cant lol just play with settings ;)
just breaking the lines with 9211 :D
holy crap tilt that is nice :)
and no can say its a cache run, there is no cache!
Wow, nice numbers tilt.
I just have to ask, are you running it as a integrated raid-0 or as pass-through software RAID?
I see according to CDM 3.0 you get 120K IOPS read, but AS SSD gives you 175K.
CDM 3.0 4KB QD32 test is single-thread, while AS SSD 4KB QD64 is 64 threads. You may have run into a CPU limit at 120K IOPS in CDM that AS SSD doesn't have.
8x25-M at 30-40K IOPS each should get you 240-320K IOPS, but you maxed out at 175K.
On the other hand, 175K IOPS @ QD 64 = 2734 IOPS pr QD.
Would you run a couple of quick IOmeter setup if i provide you with the config? (or you can make it yourself)
First will be 1 worker, 4KB random read, 4KB alligned, 1GB testfile (since you don't have cache), queue depth at exponential stepping 1-256 with 2^n stepping, 1 sec ramp time, 10-15 sec run time (since you don't have cache). It will total 9 runs and take ca 2-3 minutes.
Second will be #CPU cores (logic) workers, same access spec and test size/lenght, but QD exponential stepping 1 - {256/#workers} (say 8 "cores" makes 8 workers, each with QD 1-32, giving 6-7 runs).
Results should be fairly accurate and paint a picture of IOPS scaling. The reason for QD 1-256 is to give all SSDs a QD of 32 at the end (wich is max supported by SATA spec).
Quote:
Originally Posted by GullLars
make the file and then give it to me to run it and i will post the results and then fix them to see the specs.
Now im running on IR raid 0 64 stipe size i cant change it
im using only 1 worker.
im waiting for ur test file to run it if u promise me that u going to fix the xls file and post the results here ;)
Sure, i'll make an xls file from your raw data (.csv), and make a graph if you want.
Here's the IOmeter configs. One with 1 worker, and one with 8 workers.
BTW, i'm impressed, i thought the max IOPS from integrated RAID on the 9200-series was around 80-90K, but you've shown 175K in AS SSD. Still, that's only 22K IOPS pr x25-M, and a single x25-M can do 35-40K IOPS (4KB random read).
opened and run nothing else touched :/
So you get max 110K IOPS with 1 worker, but with 8 workers you get 198.931 IOPS, sweet.
I'll make the graphs later today, just skimmed the sheets now. I've got stuff i need to tend to :P
BTW, while i'm gone, you could try just cloning the 8 workers to 16 workers whitout changing anything else and see if you pass 200K IOPS, that would be nice ;)
What score do you get in PCmark vantage HDD with that setup? Do you beat Steve's 90K from his new WR today?
As far as I'm concerned, ne1 who hides their info does not deserve the spot they hold... Its called being a PUNK.... Futuremark should remove his score and listing for hiding is score. It just ain't right to claim world #1 and then hide the goods.... Thats like claiming that you have the fastest car in the world then telling folks when asked that oh yeah "It was so fast you couldn't even see it" Its :banana::banana::banana::banana:ing invisible..... Gimme a :banana::banana::banana::banana:ing break. Be a man and show your system details. Quit hiding for whatever reason your doing so. Your no better than Corsair if you choose to continue hiding.
LOL... chill buddy chill. Stevero has asked me several times what my config is, and I have answered him whenever I got the chance to log in and respond to my inbox...Quote:
Originally Posted by dj_ski69
My answer to him a couple of days ago - was the same as I have answered several other people who have been asking me, still running on my almost 4 year old RETAIL Areca 1231ML ( first GEN, not 2nd GEN like Stevro) with 2GB of cache and ONLY 2 ( TWO) ANS9010. And those 2 drives have been somewhat properly configured....Don't beat me up for spending the time to learn my controler and make sure I get the most out of it. And NO - not sponsored, paid for by MYSELF...
it's obvious that Stevo has been gunning for the top score for a while now. WTF is wrong with me wathcing him silently. YOU EXPECT ME TO CONFIGURE HIS CLEARLY SUPERIOR RIG SO HE CAN BEAT ME?????
WTF?
He's been getting plenty of info to figure out what he has been doing wrong. DO I REALLY NEED TO DO THAT???
Also NOT MY fault that FM can not properly detect my HW.
GET A LIFE OR :banana::banana::banana::banana: OFF
You have no idea how many times I have been asked privately in a "friendly" manner of what hardware I use, and how many times that "friendly question" turned out to be used against me to copy my configuration and beat me....I enjoy being able to figure out myself what the best possible configuration to build the fastest PC overall. I understand tho that some people prefer to just copy a proven solution and it can be helpful information.Quote:
Originally Posted by Nizzen
Give me some time on that - I have been working on a guide for the pcmarks, not quiet ready yetQuote:
Originally Posted by mbreslin
That will be a bookmark for sure! :)Quote:
Originally Posted by mike
Also - much thanks to Mike and to everyone here for all that I have learned in the past year. :yepp:
Mike, Planet, and most other folks on this forum have much more oc experience than I, it is a pleasure to learn from you all. :up:
To the person who had a go at mike for not giving his specs/secrets: If we knew *exactly* how to get the most out of pcmv, then it's just who can throw more money at it. I think 'whoever has the most money to burn wins' is horribly boring. I've enjoyed learning about my hardware and any tips for pcmv would be great. Just being handed the answer would make the whole process uninteresting for me and I'd just stop.
Edit: 'whoever has the most money to burn wins' or 'whoever has room for a SS and a wife who doesn't care' *cough* steve *cough*
I'm pretty lucky as well. My missus has let me keep a Giant PURPLE Cascade in the living room for now....:rofl::rofl::rofl:Quote:
Originally Posted by mbreslin
I bet that is an interesting conversation piece when friends or family visit :up:Quote:
Originally Posted by Expat GriZ
Yes...it HAS come up in conversation. :yepp: But the garage is stuffed to the rafters, so until I get my shed build (this summer...please!!) while I'm at work it's hiding behind 2 palm trees in a corner. She has the patience of a SAINT!! :D
So I was right.Quote:
Originally Posted by One_Hertz
Congrats on figuring out how to run the vantage bench in cache I guess.
I never did like Expat GriZ.
maybe sell me a few of those c300 for half price, to me of course, and get enough for phase mbreslin?:hump:
@Expat - that is a beautifull machine!
@mbreslin - yes - my wife is a saint also.
By the way - my SS was much less than you might think - contact xtremesystems member - sdumper for details.
Quote:
Originally Posted by One_Hertz
What's the point that you are trying to make here? That you are right?
When I had spoken to a friend of mine ( who is a very active member in the storage forum ) before posting results, I had asked, should I beat Stevero in the manner he is running, or should I go all out and cache the FIRST subtest ontop of it?
His response was - go all out, and show how it is done.
Sorry if I displeased you by doing so.
From you negative undertone in your comments, it appears that you disagree with me using my RAID controler to it's max capabilities. Why?
My primary use form my Areca is video editing. I need to use all the system memory that is available to me. I cannot afford to create a cache drive based on system memory. A raid controler with onboard cache, is a solution with real world benefits and advantages. Want me to cripple my rig even more as it already is???
People have been using software based cache drives in PCMarks for a while now. Some intentionally, others accidently. I can clearly see the real world benefits from running software based cache drives on your system memory, but for me it is not an option, and I prefer to use a DEDICATED hardware solution for my and this purpose.
Do you also complain when people are using Intel's onboard raid controler, or some of Adaptec's controler where the driver smartly creates a cache drive using your system memory? OR some clowns who ontop create multiple small arrays, software stripe these again, thus creating a BIG software based ramdrive with the sole purpose of gettting a top score in PCMarks?
If you don't like what I am doing just ignore me. But PLEASE don't be-little me.
Afaik I still have the highest HDD suite overall score without any RAID controler caching involved, to appease even folks with your viewpoint.
Week before last I figured out how to cache the first subtest - it is in my latest results also :)
If you like to benchmark the speed of your cache then go ahead. Those scores don't have much to do with the actual speed of the array, but if that is fine with you then congrats. To each their own. Will take your advice and ignore you from now on.Quote:
Originally Posted by mike
And yes, I would say the same to anyone a using software ramdisk or even to the ICH10R results where it shows 1200mb/s for one of the tests when caching is turned on (media test I think).
It doesn't???? Really? Sorry had no ideaQuote:
Originally Posted by One_Hertz
SO all the snappy performance that I thought I had all those years while using my Areca 1231ML's array in real world useage where just in my mind - and not REAL???
Damn I must be easy to fool
Save you snippy grats - not interested in them...Quote:
Originally Posted by One_Hertz
Agreed to each their own. :up:
sorry mike but your 1231 is total crap!! :rofl:
and according to this statement its no better @ real apps :rofl:Quote:
Originally Posted by GullLars
mike your 1231 is just a stinkin pos!! :up: :D
Makes me want to get a 1231 :x
Edit: Also, since mike and steve both cheated, and planet worked his ln2 magic for corsair, I'm claiming first place.. (nobody tell nizzen)!
:D
nowhere in that do i see gullars saying that it is no better at real apps. not once does he say that. he just says your taking advantage of cache performance. not that it is bad and he is not belittling you napalm. neither does he say the 1231 is a "stinking pos".Quote:
Originally Posted by GullLars View Post
If you insist on spamming one application, try making a "multi-thread" batch setup, make one batch launching 10 other, and then have those 10 launch N instances of the program each, like 30 for a total of 300. Should be interresting to see if the QD skyrockets.
I stilll suspect the ML1231 reads every FF instance afte the first 2-10 from cache, and thereby get cache speeds instead of disk speed, so you'd basically be benching the RAID controllers caching performance.
he just points out a fact that when you are doing YOUR little test you are doing that it is reading from cache.
which it is.
Quote:
Originally Posted by NapalmV5
:p: truly sux that thing...
As passionate as I have responded to some of the points being made, I do have to say that I do see some validity in some of the points in respect to raid controler caching. For sure there are two sides to each coin.
The caching on the controler for sure can help disguise some of the shortcomings of the drives being utilized, but where shall the line be drawn??? What is real world useabale and what is not???
Do you want to turn off the cache on old spinning HDDs, to assess the performance, disable cache on SSDs like the OCZ Vertex etc. etc. Where do we draw the line?
For me the line is drawn when an artifical cache is created on system memory, which steals away my useable memory for my applications.
For someone like One_Hertz the line is drawn earlier. But I see a test like Vanatage as a benchmark that assesses your entire storage subsystem, which in my point of view includes the Raid Controler and it's full capabilities. I don't think anyone would want to have to turn off the cache on each indivudal drive...so the line gets murky...
people are gonna run cache on PCMV, period. there is now way around it. as long as it is a BOOTABLE array i see no problem with it. we are measuring systems, as a whole. if mike uses his system to manipulate large files (video) then he is definitely reaping the benefits of his cache with daily usage. i have to agree that the BS with the non-bootable arrays being off limits. but i see nothing with mike or steve or anyone for that matter using their systems to their full potential. they put cache on these controllers, and the very devices themselves, for a reason. to USE it.
I have held #9 and 10 and 13 now and everything inbetween, and if you believe that my little bitty 512 mb of cache had nothing to do with it then your an idiot. every single person on that list has cache on their system SOMEWHERE that is being used. like he says, you want my cache on my eight vertexes off too?
@napalm---Keep yoiur silly grandstanding over in your own thread. dont bring an argument from your thread over here into something unrelated. of course your cache is being used. we are all acknowledging that as well.
also, quoting gullars and taking his statements out of context in an entirely different thread is bull:banana::banana::banana::banana:. last i checked gullars is a pretty respected person on this forum so act your age.
have a nice day.
To present my personal point of view a bit further - I believe anything that is sustainable is fine.
For example - hooking up a single old HDD that can write at 50mb/s to a 2GB arc 1231. If you write 2GB of data to that array, the speed of writing will be 1,500mb/s and it will "finish" in 1.33 seconds. If you write 4gb to that array, it will take 41.33 seconds because you run out of cache. I do not believe saying that this particular array can write at 1,500mb/s is valid, although it technically CAN for some period of time. The sustainable speed is 50mb/s. The correct thing in my mind would be to say that this is an hdd with a write speed of 50mb/s + 2gb of cache...
I'm ignoring all of you for the moment - doing real work I want you to know.
I just luberexed my socket and conformal coated the board.
UPS says they have a big wooden box for me - arriving tomorrow :up:
kneaded eraser makes the microwave spark - tough it out microwave!
Making synthetic winter in your house is cheating, anyone who wants to do subzero for benching should have to move to somewhere with subzero ambients. Any pcmv scores from you will not be allowed in the guiness book of world records. You have been warned.
lol..good luck steve-o!
helping conformal coating dry
http://img62.imageshack.us/img62/706/dsc03290v.jpg
kneaded eraser as insulation
http://img687.imageshack.us/img687/1006/dsc03293r.jpg
http://img707.imageshack.us/img707/4120/dsc03294yc.jpg
I want to make it clear, I do not diss the RAID controllers with cache, or the very real performance boost it brings, but i question the point of things when the owners of these cards start to do things like opening 100's of instances of the same program and timing it.
I agree with Computurd, for a PCmark Vantage HDD score to be valid in my book, it has to be bootable, and not just be artificially boosted by running purely in cache, like most normal loads never would (they could run partially, or mostly, but few loads would run purely in cache).
Like others noticed, in the quote from last page where i note the 300+ FF instances loading running in cache, i actually suggested a way to possibly get better results from making it multi-threaded by running many batch-files in parallell. And seriously, i'd like to know if it makes an impact on QD.
Mike, the ca 6K points you have left to harvest, do some come from L2N or other means of extreme cooling you currently don't have hooked up?
EDIT: Nice pics steve, i wait in anticipation to see how it evolves :)
didn't think so. Why would anyone want to diss 'em ;)Quote:
Originally Posted by GullLars
Most of the gain would come from using LN2; also by using a different CPU and or maybe a new controler :D :rofl: not sure what kind yet :DQuote:
Originally Posted by GullLars
Also interestingly enough; HDD performance will also get a nice boost; just from running very high CPU clocks. I expect to get about 4k worth of points from system performance, and about another 2k worth or overall points coming just from HDD subtests.
I know SteveRo is gonna make me bench Vantage yet again in the next few days...argh...hopefully UPS will be on strike for the next few days....Quote:
Originally Posted by GullLars
Aw gee, schucks...& here I thought we were getting on fine. :confused: Didn't you and your wife go to "cold councelling"?? :D (highly recommended):yepp:Quote:
Originally Posted by mbreslin
Tell you what, you hand over all those crucial drives, you have too many anyway, and we can be pals again.!!!:rofl::rofl::rofl:
2 both Mike, Steve & anybody else joining this battle...well done. :up: I don't know enough about ram disks, arrays...iops, controllers....etc etc. I just enjoy the FRIENDLY competition.
Be Xtreme......and keep benching!!! :up:
It makes me feel a little better that you don't already have them in the bank this time! :)Quote:
Originally Posted by mike
Take your time on the come back - let me bask in the glory a little longer this time will ya? :rolleyes:
What do you mean by conformal coating?Quote:
Originally Posted by SteveRo
The kneaded eraser insulation looks like you did a good job. You might want to stick some sort of insulation (if possible) in the first ram slot as well.
conformal coating - like nail polish but special made for electronics - http://en.wikipedia.org/wiki/Conformal_coating
What would you recommend for the 1st memory slot - what do you use?
I haven't been able to find that stuff or even liquid electrical tape anywhere up here.Quote:
Originally Posted by SteveRo
You could take a piece of a shop towel and shove it down the DIMM slot. Or a very thin slip of insulation. All my DIMM's are full so I don't use anything.
Lowf - How low T do you go with your setup?
Also - liquid electrical tape you can get at Walmart in VA.
You can get conformal coating here - http://www.crazypc.com/products/conf...ng-934900.html
Edit - oops, out of stock.
When my single stage was working properly evap temps under load were around -30C IIRC.Quote:
Originally Posted by SteveRo
How conductive are shop towels - I suspect if they get wet - probably very conductive.
I suppose I need to keep an eye on that first memory slot -if it looks like I am getting any moister I'll need to do something - maybe vasoline - I hate the grease
I dont think you will have any issues at single stage temps. Just throw some painters tape over it if your worried. I wouldnt recommend putting anything in the slot.
Thanks much Mr Planet - sounds like a plan - will do.
Are we going to see some more pcmv scores from you?
You don't need anything in the 1st DIMM slot unless 1) you are sloppy pouring LN2 2) you are running below -150C.
Thanks gomeler.