Register a SA Forums Account here!
JOINING THE SA FORUMS WILL REMOVE THIS BIG AD, THE ANNOYING UNDERLINED ADS, AND STUPID INTERSTITIAL ADS!!!

You can: log in, read the tech support FAQ, or request your lost password. This dumb message (and those ads) will appear on every screen until you register! Get rid of this crap by registering your own SA Forums Account and joining roughly 150,000 Goons, for the one-time price of $9.95! We charge money because it costs us money per month for bills, and since we don't believe in showing ads to our users, we try to make the money back through forum registrations.
 
  • Post
  • Reply
Agreed
Dec 30, 2003

The price of meat has just gone up, and your old lady has just gone down

Wedesdo posted:

Hum my new h100 is coming. Tempted to jack my 2500K up to 1.42V and see if I can get 5 GHz.

What's the max safe voltage for a sandy bridge again? I heard ~1.45 is okay on water?

You can hear a lot of things. Yes, lower temperature is important because heat adds resistance and that can cause unpredictable behavior and early part degradation (perhaps even more important when said part is 32nm in size :)). However, operating voltage is its own concern separate from as well as in conjunction with temperature. High voltage + high temperature is outright dangerous, but high voltage can be dangerous even without high temperature.

One thing in your favor if you do want to try some crazy poo poo overclocking, you could just buy the overclocking insurance and if you are able to keep temps down, at least it won't have a burnt mark on it if it fails since it'll just be a failure of the microarchitecture or something else under the heatspreader. But it's been mixed as far as what's "safe" for Sandy Bridge anyway, some people have been running the chips near 1.5V for some time with no issues, but others have definitely fried their processors. 1.38V is considered safe, but the VID specifies... in its peculiar way... around 1.5V. That just happens to seem really damned dangerous in practice. Intel hasn't been super helpful on that front.

Your call.

Adbot
ADBOT LOVES YOU

Factory Factory
Mar 19, 2010

This is what
Arcane Velocity was like.

Wedesdo posted:

Hum my new h100 is coming. Tempted to jack my 2500K up to 1.42V and see if I can get 5 GHz.

What's the max safe voltage for a sandy bridge again? I heard ~1.45 is okay on water?

It is 1.38V if you expect the chip to outlast its warranty. If you're gonna jack it up higher, get the overclocking insurance. $20 for an i5-2500K and you can volt it however high you like.

Wedesdo
Jun 15, 2001
I FUCKING WASTED 10 HOURS AND $40 TODAY. FUCK YOU FATE AND/OR FORTUNE AND/OR PROBABILITY AND/OR HEISENBURG UNCERTAINTY PRINCIPLE.

Well, it's at 1.38V and 4.8GHz right now. I shouldn't get too greedy, right?

Eletriarnation
Apr 6, 2005

People don't appreciate the substance of things...
objects in space.


Oven Wrangler

Wedesdo posted:

Well, it's at 1.38V and 4.8GHz right now. I shouldn't get too greedy, right?

That's really good from what I've read, so if you kill it and get a replacement I wouldn't expect to be able to take the replacement that high.

hobbesmaster
Jan 28, 2008

Perhaps there is a better thread for this, but is there any news on general availability of Knight's Corner/Intel MIC?

Star War Sex Parrot
Oct 2, 2003

hobbesmaster posted:

Perhaps there is a better thread for this, but is there any news on general availability of Knight's Corner/Intel MIC?
Probably not any time in the near future, if I had to guess. It's dependent on Intel getting their 22nm 3D tri-gate process perfected, which Ivy Bridge seems to be the first commercial guinea pig for.

hobbesmaster
Jan 28, 2008

Star War Sex Parrot posted:

Probably not any time in the near future, if I had to guess. It's dependent on Intel getting their 22nm 3D tri-gate process perfected, which Ivy Bridge seems to be the first commercial guinea pig for.

Thats unfortunate, was hoping for a 75W supercomputer. Video cards take so much power...

edit: Upon looking that up again, did I imagine 75W?

Alereon
Feb 6, 2004

Dehumanize yourself and face to Trumpshed
College Slice
Yesterday VR-Zone published some leaked Ivy Bridge CPU benchmarks, the Core i7 3770K (clocked the same as a Core i7 2700K) beat the Core i7 2600K by ~8.9%, which is about a 6% clock-for-clock improvement when you consider the slight clockspeed advantage. At the same time power usage is going down by ~19%, and we haven't even seen the graphics performance increases and overclocking potential yet.

tijag
Aug 6, 2002

Alereon posted:

Yesterday VR-Zone published some leaked Ivy Bridge CPU benchmarks, the Core i7 3770K (clocked the same as a Core i7 2700K) beat the Core i7 2600K by ~8.9%, which is about a 6% clock-for-clock improvement when you consider the slight clockspeed advantage. At the same time power usage is going down by ~19%, and we haven't even seen the graphics performance increases and overclocking potential yet.

My hope is that we can hit 5ghz on air pretty easily with IVB. I think 4.5ghz was pretty doable for SNB, so can we get another 500mhz out of IVB?

Alereon
Feb 6, 2004

Dehumanize yourself and face to Trumpshed
College Slice

tijag posted:

My hope is that we can hit 5ghz on air pretty easily with IVB. I think 4.5ghz was pretty doable for SNB, so can we get another 500mhz out of IVB?
That's my hope as well, and it doesn't seem particularly unreasonable if you use a top-end air cooler.

in a well actually
Jan 26, 2011

dude, you gotta end it on the rhyme

hobbesmaster posted:

Thats unfortunate, was hoping for a 75W supercomputer. Video cards take so much power...

edit: Upon looking that up again, did I imagine 75W?

Late 2012. Expect power consumption and theoretical floating-point performance similar to Kepler.

hobbesmaster
Jan 28, 2008

theclaw posted:

Late 2012. Expect power consumption and theoretical floating-point performance similar to Kepler.

If nVidia's roadmaps are to be believed then Maxwell should blow that out of the water a few months later?

IPP/MKL integration would be much easier to deal with than CUDA however.

feld
Feb 11, 2008

Out of nowhere its.....

Feldman

Contact of mine @ SuperMicro claims to have IvyBridge Xeons in his lab "and drat they're quick". Can't validate it one way or the other but it's something I like to dream about :allears:

movax
Aug 30, 2008

feld posted:

Contact of mine @ SuperMicro claims to have IvyBridge Xeons in his lab "and drat they're quick". Can't validate it one way or the other but it's something I like to dream about :allears:

They're fast (and cool) :ninja:

Endymion FRS MK1
Oct 29, 2011

I don't know what this thing is, and I don't care. I'm just tired of seeing your stupid newbie av from 2011.
I went ahead and bought an Intel DP67-BG3 board, since they were on sale last week at Newegg. After buying it I decided to go ahead and wait for Ivy Bridge rather than pick up an i5-2500k based on what I've read. I saw that Ivy Bridge is backwards compatible with P67 and Z68 boards through a bios update, so that relieved me somewhat on if I had wasted money on a new board. My question is will I be able to do the update without a Sandy Bridge processor in there in the first place? Or should I just return it and grab a new board once the new line comes out?

incoherent
Apr 24, 2004

01010100011010000111001
00110100101101100011011
000110010101110010
IB is unique as one mfg is shipping bioses with IB cpu support right now. However there is a caveat to IB support: It is not 100% guaranteed. If your motherboard mfg cheapen out on the bios chip, you are SOL.

Intel is not above any of this as well. They've shipped motherboards that wouldn't support the next revision of a processor (conroes->wolfdale).

I'd recommend getting the 2500k and keeping the board. Its a very respectable board.

Alereon
Feb 6, 2004

Dehumanize yourself and face to Trumpshed
College Slice
I'd strongly recommend you just return the board and buy whatever is a good deal when you're ready to build. Buying parts on sale that you don't plan to use immediately almost always means you lose money in the long term. You give up some of your warranty term, too.

future ghost
Dec 5, 2005

:byetankie:
Gun Saliva

incomprehensible posted:

IB is unique as one mfg is shipping bioses with IB cpu support right now. However there is a caveat to IB support: It is not 100% guaranteed. If your motherboard mfg cheapen out on the bios chip, you are SOL.
Crossing my fingers with my Gigabyte board. If it's not supported later on, I can deal, but they have a thorough history of supporting new chips well-beyond when they should. I still have a 965P board from them that runs 45nm quads with a BIOS update. The last socket 775 Intel board that I owned (DG31PR) would run 45nm chips, but it never received a microcode update, so it was convinced I was running a bunch of Pentium-III's. That board was strange in alot of ways though.


Endymion FRS MK1 posted:

My question is will I be able to do the update without a Sandy Bridge processor in there in the first place? Or should I just return it and grab a new board once the new line comes out?
Srebrenica Surprise posted this when you asked this question in the parts megathread:

quote:

I remember reading something (probably in the IB thread) saying Intel boards wouldn't be compatible with IB like other boards are because they don't have enough space for a BIOS flash or something. Does anybody know if this is correct?

In response to that comment, read this post on vr-zone:
http://vr-zone.com/articles/the-upgrade-path-to-ivy-bridge-might-be-blocked-by-changes-to-uefi/13513.html

Basically, you might be able to drop an Ivybridge chip in it, but it's unlikely that you'd even be able to attempt it without another SB chip to run a BIOS flash (and even then a BIOS update may not allow it as Intel boards have less flash memory to work with, possibly preventing the required UEFI update). Not to mention, if the board is DOA you wouldn't be able to find out within the return window.

Return it as Alereon suggested and wait.

future ghost fucked around with this message at 19:28 on Feb 12, 2012

Endymion FRS MK1
Oct 29, 2011

I don't know what this thing is, and I don't care. I'm just tired of seeing your stupid newbie av from 2011.
Ok, thanks guys. Thats what I get for buying on impulse I suppose.

mobby_6kl
Aug 9, 2009

I know I've previously said here that Sandy doesn't offer enough benefit over C2Q processors for me to bother and that I'd wait for Ivy, but now that it's almost here... are there any details on Haswell? If there was any bottleneck in my setup, I thought it was mostly the platter storage (system drive just died) and graphics (stock fan just died). So if I replace both of those, I'd buy myself quite a bit more time. Have anyone given any thought to the next "tock"?

Endymion FRS MK1
Oct 29, 2011

I don't know what this thing is, and I don't care. I'm just tired of seeing your stupid newbie av from 2011.

mobby_6kl posted:

I know I've previously said here that Sandy doesn't offer enough benefit over C2Q processors for me to bother and that I'd wait for Ivy, but now that it's almost here... are there any details on Haswell? If there was any bottleneck in my setup, I thought it was mostly the platter storage (system drive just died) and graphics (stock fan just died). So if I replace both of those, I'd buy myself quite a bit more time. Have anyone given any thought to the next "tock"?

Isn't this just a cycle where you constantly wait till the next tick or tock then decide to wait until a tock or tick instead and repeat? Forgive me if that sounds a bit crude or whatever. I did have that thought though too (running a C2Q now as well), then I decided to bite the bullet and go for Ivy Bridge. Sooner or later you'll have to actually upgrade instead of keeping on waiting for the next development

Gwaihir
Dec 8, 2009
Hair Elf
I dunno, I was running an overclocked Core2 (nothing extreme, probably 3.2ish ghz) when the first Nehalem i7s came out, and the move to an i7-920 OCed to 3.8 ghz was pretty damned impressive, and it's been a fine machine since November 2008 when I put it together. I put in an X-25m later on, and that was an even better upgrade, but the core2s, even OCed ones, are preeety damned creaky at this point. All of our work machines are STILL core2 based (Government, etc etc). My workstation happens to have a 240 gig Vertex2, and even with that, it is not close to my i7-920 in terms of responsiveness on the desktop. Obviously can't say for gaming on the work machine, but yea. With an SSD they're certainly not bad machines, and I would do an SSD upgrade before the processor, but you're still going to be pretty damned happy with a Sandy Bridge or Ivy Bridge upgrade, even if your Core2 is OCed to hell.

Beelzebubba9
Feb 24, 2004

mobby_6kl posted:

I know I've previously said here that Sandy doesn't offer enough benefit over C2Q processors for me to bother and that I'd wait for Ivy, but now that it's almost here... are there any details on Haswell? If there was any bottleneck in my setup, I thought it was mostly the platter storage (system drive just died) and graphics (stock fan just died). So if I replace both of those, I'd buy myself quite a bit more time. Have anyone given any thought to the next "tock"?

There are a bunch of architectural details on Haswell available, but the general gist is that Intel is targeting portables even more aggressively with its design. From what I remember, Intel is pushing to get Haswell down to a 10W thermal envelope, so a lot of the optimizations will be in that direction. I don't expect Haswell to be drastically different from IVB - just a continued evolution of the concepts Intel has already introduced.

That said, I'd base any upgrade decisions based on the ability of your current rig. I certainly wouldn't put off a CPU upgrade by 18 months based on a 'what-if' if your Core 2 is getting long in the tooth.

Standish
May 21, 2001

mobby_6kl posted:

are there any details on Haswell?
This is really cool, more of a big deal for servers than desktops though.

mobby_6kl
Aug 9, 2009

Well I'm definitely well aware of the "there's a better model in x month" thing, so I'm mainly trying to judge the upgrade based on how my current processor satisfies the performance demand, which is pretty well. I simply don't recall wishing for a faster one very often (maybe when compressing large amounts of data or rendering something, but I don't do that too frequently).

I've actually seen the transactional memory article on Arstechnica but didn't really read it properly or connect that it'll be in the next architecture. It does look like something that could make a real difference on utilization of whatever ridiculous amount of cores that these processors will have. It's pretty exciting, really, even if it doesn't instantly take hold on the desktop, just from an amateur programmer's point of view.

hobbesmaster
Jan 28, 2008

Standish posted:

This is really cool, more of a big deal for servers than desktops though.

FMA will also be pretty cool for some applications.

PUBLIC TOILET
Jun 13, 2009

So if Gigabyte has completely dropped the ball on their motherboards when it comes to the Vdroop issue, which manufacturer has worked around this issue the best? I'm probably not going to build a new system until Ivy Bridge arrives, but I've grown accustomed to Intel reference boards (I don't do much overclocking at all and I have a habit of keeping the vital components to one manufacturer). However, I'm willing to jump ship in that regard and consider Asus/Asrock or MSI. I've had ASUS boards before but that was a while back and the only thing I notice about them now is the high price. But I suppose that's where Asrock/MSI come in.

I don't even know how Intel handles the Vdroop issue with their own motherboards. Maybe they do a decent job?

redeyes
Sep 14, 2002

by Fluffdaddy

COCKMOUTH.GIF posted:

So if Gigabyte has completely dropped the ball on their motherboards when it comes to the Vdroop issue, which manufacturer has worked around this issue the best? I'm probably not going to build a new system until Ivy Bridge arrives, but I've grown accustomed to Intel reference boards (I don't do much overclocking at all and I have a habit of keeping the vital components to one manufacturer). However, I'm willing to jump ship in that regard and consider Asus/Asrock or MSI. I've had ASUS boards before but that was a while back and the only thing I notice about them now is the high price. But I suppose that's where Asrock/MSI come in.

I don't even know how Intel handles the Vdroop issue with their own motherboards. Maybe they do a decent job?

I had a Skull Trail a week ago and they did in fact have a Vdroop compensation setting in the BIOS. I never tried it and don't know how well it works.

DaNzA
Sep 11, 2001

:D
Grimey Drawer

movax posted:

They're fast (and cool) :ninja:

Can't wait for the extreme edition 8/10 core IVB where they can keep all the cores enabled while still running within a reasonable thermal envelope at a decent clock speed.

Alereon
Feb 6, 2004

Dehumanize yourself and face to Trumpshed
College Slice
EXPreview has some details and graphics benchmarks of the upcoming Ivy Bridge Core i5 3570K. HD Graphics 4000 is between 30-85% faster than HD 3000 in real games, up to more than twice as fast in 3DMark.

VR-Zone is also reporting that the previously announced delay for Ivy Bridge will only apply to dual-core mobile chips. Desktops and quad-core mobile chips will launch at the beginning of April as planned.

Agreed
Dec 30, 2003

The price of meat has just gone up, and your old lady has just gone down

Alereon posted:

VR-Zone is also reporting that the previously announced delay for Ivy Bridge will only apply to dual-core mobile chips. Desktops and quad-core mobile chips will launch at the beginning of April as planned.

That both makes more sense from a logistics standpoint (clearing those out to oems must be a pain in the rear end, and they are good for nothing afterward) and, surprisingly, from a "let's not squander goodwill with end users and organizations" standpoint for the performers in the lineup.

Alereon
Feb 6, 2004

Dehumanize yourself and face to Trumpshed
College Slice

Agreed posted:

That both makes more sense from a logistics standpoint (clearing those out to oems must be a pain in the rear end, and they are good for nothing afterward) and, surprisingly, from a "let's not squander goodwill with end users and organizations" standpoint for the performers in the lineup.
There was also a pretty long delay between the introduction of Sandy Bridge and its appearance in less expensive laptops as the Arrandale stocks declined. Granted we're always going to see a time period where last-gen stuff is sitting on the shelves, but hopefully this plan will give them time to discount Sandy Bridge notebooks and get them off the shelves before replacing them with Ivy Bridge products, for a faster transition once the IB i5s actually hit in June.

Factory Factory
Mar 19, 2010

This is what
Arcane Velocity was like.
Anyone have more intel (har har) on Intel's Near Threshold Voltage efforts than AnandTech?


clicky

The thing is Claremont, derived from an original Pentium and built on 32nm processes with NTV research enhancements. It idles at 3 MHz @ 280mV, and scales all the way up to 915 MHz @ 1.2V. Its lowest power state draws only 2mW. That teeny solar panel is sufficient for the chip to run Windows or Linux.

It looks pretty rad.

sleepy gary
Jan 11, 2006

Intel loving owns. They're doing some really excellent stuff. I hope someday I will get a chance to work on some of that.

JawnV6
Jul 4, 2004

So hot ...
More ISSCC coverage: http://www.eetimes.com/electronics-news/4236562/Intel-gives-deeper-look-into-Ivy-Bridge?cid=NL_EETimesDaily

SRQ
Nov 9, 2009

If I understand this right, they took the original pentium design and die shrunk it + added new instruction sets?

Longinus00
Dec 29, 2005
Ur-Quan

Factory Factory posted:

Anyone have more intel (har har) on Intel's Near Threshold Voltage efforts than AnandTech?


clicky

The thing is Claremont, derived from an original Pentium and built on 32nm processes with NTV research enhancements. It idles at 3 MHz @ 280mV, and scales all the way up to 915 MHz @ 1.2V. Its lowest power state draws only 2mW. That teeny solar panel is sufficient for the chip to run Windows or Linux.

It looks pretty rad.

Is that solar panel really sufficient for it to run windows or just putter around in one of the low energy P states?

Alereon
Feb 6, 2004

Dehumanize yourself and face to Trumpshed
College Slice
Man, I thought EEtimes was good but every other sentence in that article is some sort of inexcusable factual error.

SRQ posted:

If I understand this right, they took the original pentium design and die shrunk it + added new instruction sets?
Essentially yes, that's what the Intel Atom is. There are more changes than that, but Anandtech describes it as "imagine if Intel could go back and remake the original Pentium processor, with everything its engineers have learned in the past [20] years and build it on a very small, very cool [...] manufacturing process." The downside is that in-order designs are obsolete and inefficient, so while the CPU can operate with very low power, a more efficient out-of-order processor like those offered by ARM and AMD are vastly better. Anandtech published benchmarks of the new Qualcomm Krait cores today, it's more than twice as fast as any existing smartphone processor. Intel's Silvermont processors, due in 2013, use a new out-of-order architecture and other changes that should make them more competitive, such as an integrated memory controller to reduce latency and improve bandwidth. Current Atoms use a memory controller built into the Northbridge chipset, and while that is on the same die as the processor, it still incurs the same performance penalty as non-integrated designs since data still has to traverse a Front Side Bus.

movax
Aug 30, 2008


quote:

Specifically the largest die includes four x86 cores and a large graphics block. It can be chopped along its x- and/or y-axis using automated generation tools to create versions with two cores or a smaller graphics block.

This owns. I've only ever done chip design small enough for MOSIS to handle.

Adbot
ADBOT LOVES YOU

Factory Factory
Mar 19, 2010

This is what
Arcane Velocity was like.
By the by, anyone concerned about competition and innovation in light of AMD's restructuring might take some solace if you read up on Joseph Schumpeter's theories of monopoly and innovation. In a nutshell, he said that monopolies can drive innovation because they have large amounts of capital which can be invested into novel research. As semiconductor design is a highly investment-driven industry, it's a perfect place for such a market dynamic.

  • 1
  • 2
  • 3
  • 4
  • 5
  • Post
  • Reply