That is why I said it isn't exactly a FIFO. ;-) But the effect is similar in that data is stored on the ram in the drive. You can see benchmarks where if the file being written is very large and comes from a simple copy operation, the drive becomes an issue. But in the case of spice, the data is being computed rather than copied, so if the circuit is anything but trivial, the write demands shouldn't be the limiting factor.
BTW, as a linux user, I don't defrag. Using ext4, your files begin life being fragmented.
Benchmarking has shown quad channel isn't useful at the moment.
Xeon is not expensive if you get the E3 1200 series.
Use a Supermicro mobo and use the error detecting correcting ram. I can swing a dead cat and it would land on a local system builder who could put that together if you can't do it yourself.
The E3 1200 chips are Haswell chips that have Xeon tweaks. You get all the virtuallization. They handle error correcting memory of the UNBUFFERED variety. That confuses a lot of people. Because it uses unbuffered RAM, it is limited to 32Gbytes. Xeons that use registed memory can use more ram, but that ram is slower.
Dell has the E3 1200 V3 in some of their Power Edge products. I have no idea if they are good PCs since I build my own.
The Xeon product line is all about stability. No overclocking. They use ECC, which some say is slower. [I don't know.] If you are seriously going to do a ram disk (dumb idea), you would want the ECC. For software RAID, you should have ECC. I give Dell credit for at least using a Supermicro mobo, since some of the Asus mobos don't use ECC correctly.
The bad news is RAM prices are up for some reason.
Hard drive caches are huge and grab nearly entire cylinders of data.
Fragmentation WAS a BIG problem when drives were physically bigger, and the heads moved slower, and the data trailed out onto longer lineal strings. Seek times, and head transitions all had a huge cost on the speed of the data retrieval as well as the longevity of a given drive.
This is no longer the case. The heads only transit less than an inch across its entire gamut of travel. That happens a lot faster these days.
That and the caching makes it an issue of the past. I see hard drives getting hit every second of every day, and they last for years.
I seriously doubt they exhibit any serious access time hinderances due to 'fragmentation', regardless of the OS.
EVGA did the X79 series pretty good, and they are screaming on the newer chips too.
Oh, and one most certainly CAN OC a Xeon. It is all about the MOBO you mount it on.
And my philosophy on ECC is... I have it on my OLD AMD X2. It costs more, but there has never been a failure to catch.
If a CPU has upwards of 3 billion transistor elements in it, and they breeze along without ever latching, then I think we can all safely use non-ECC type RAM, and never see an issue.
I think it is even hard to find any more... Maybe even in commercial circles, because it use is being phased out of industrial builds even. They'll probably kill the lines soon enough, if the market wont support the extra costs of making (and testing) them.
If you are not going into space, and would have a worry of a gamma ray event, then you do not really need it, particularly when millions of PC chug along without it just fine.
I rarely see full bore hard latch ups on PCs these days.
I do still see power supplies with 'hard start' issues. I have a damn TV that only turns on when it is real cold. It will stay on after that
24/7/365, but most times, one of the 12V supplies latches on start, and the thing falls into limbo and won't start up. They gave me my money back on it. I should just go buy a 12V supply, and hang it onto the pins. One of these days, I'll do it.
That would be a big change-- the last time I heard ME on the subject he said that from his perspective AMD had no reason to exist (and the way they're going... )
Best regards, Spehro Pefhany
--
"it's the network..." "The Journey is the reward"
speff@interlog.com Info for manufacturers: http://www.trexon.com
Embedded software/hardware/analog Info for designers: http://www.speff.com
The X79 chip really doesn't handle ECC. That is why I keep saying to use a Supermicro mobo.
Read this thread:
formatting link
Regarding RAID and ECC:
When I built my Xeon desktop, I simply bought the Kingston ECC with the Intel qual. Yeah, the intel qual cost me an extra $15. ECC added about $50 to the total cost.
Today's fast RAM, ECC capable or not, is quite robust.
You are retarded to think otherwise, and a thousand OC Youtube videos further cement that fact. Some of them by a factor of three. Your claim is bogus.
Your mindset, and yardstick are what is junk, "pro-box" utter dimwit mentality horseshit.
Supermicro has you and a couple others here cowed with their utter bullshit.
Look at the benchmark sites. Pretty easy to see. One does NOT see many "super micro based "high end set-ups being shown, and it is not like they do not exist in the database, they are just a lot further down the list.
A real joke for those of us who actually know. The benchmarks do not lie, and ther are iterations with ECC as well.
Quite straightforward that your "big box store" mindset is convoluted compared with reality.
Consumers are not dumb and they drive what those stores stock. Oops... There is a *huge* flaw in your logic.
John Larkin - as a card carrying egomaniac - seems every correction or eluc idation as an attempt by the person posting the correction to establish how much smarter they are than everybody else.
The concept that bad information misleads people, is consequently potential ly dangerous (or at least time wasting) and should be corrected to make eve rybody's lives easier, is not one that he can get his (somewhat inflated) h ead around.
From his point of view, it's all about status, not about making the world r un better.
Neither is implied by the web-page posted (which comes from the University of Bolton in the UK - one of the "new universities" produced by re-badging the old technical colleges originally introduced to handle the more academi c bits of apprentice-ship training).
Not that John Larkin does design - he specialises in persistent improvisati on, which works nearly as well, if you are as good at it as John Larkin is.
I don't know who ME is. I would agree from a stock investment perspective. They used to play catchup with Intel every three or four years and if you paid attention you could make some real bucks when their stock price jumped up and down like a yo-yo. But once they got a full process generation behind and the architectures all reached a plateau, there is not much they could do to get ahead anymore. Once Intel got so far ahead in process geometry it became a matter of accounting more than technology. Intel makes chips at a fraction of the cost of the AMD chips... game over.
But that doesn't mean AMD can't have an emphasis in an area that gives them a niche market where they lead technically.
I must agree. Once some time ago my then buddy and i wrote the test data acquisition and analysis suite for a discretes test lab. You cannot imagine how flabbergasted i was to find out that most of that code was/is still in daily production use over 20 years later, untouched. They had/have personnel that can, as they have extended the suite to use new instruments but the core is untouched.
They provide one very valuable service, they help keep Intel more or less honest instead of letting them become a monopoly. Do you see any value in that? Dollar for dollar and MHz core for MHz core AMD chips have more relative FP performance and less integer performance. For a complex of reasons i usually buy AMD.
ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here.
All logos and trade names are the property of their respective owners.