Anandtech Review of Ivy Bridge is Up now! Other sites added

[BD231]

...

Definitely ready for your extreme OC thread.

I assume you're referring to Quick Sync? That seems to be the one area of substantial benefit with the new IGP. I'm just not sure how it will benefit me.

What do you mean about the post processing?

When I first read about Quick Sync the reviews seemed to imply it was doing it's own form of compression and there were quality issues with what was rendered. I thought initially that you could just use it with any encoding process you would normally use as long as it was supported. If that's the case, why would it affect quality?

Right now I'm using AnyDVD/CloneDVD and I'm wondering if it would benefit me there. Since I'm not doing a lot of gaming or a ton of video rendering, if I can get away with not getting a discrete GPU then IB would represent a savings. Supposedly it also beats any current discrete card for encoding so even if you have a discrete card for gaming, etc.. this would still represent a performance increase when using Quick Sync?

Sorry, just trying to understand all this and the practical applications. One concern though is that it still doesn't support 24p properly as far as I know.

Thanks,

Anthony

Post as in gaming.

http://semiaccurate.com/2012/04/23/comparing-the-ends-of-the-spectrum/

 

[exar333]

The water-cooling fans must be excited for IB. Its been SOOOO long since a solid WC rig made a big difference compared to a higher-end air solution. Its really been since the C2Qs since WC rigs really allowed for a appreciably-better OC ceiling.

 

[AFQ]

It isnt that hot. I mean it is hotter than Sandy Bridge when we compare clock to clock or voltage to voltage but overall its fine. Its fine even on air cooling. See this, 4.64 GHz and load temps are 75 with Venomous X: http://www.maximum-tech.net/intel-core-i7-3770k-3-5-ghz-ivy-bridge-review-12192/

though IVB seems to scale really well with cold. We are already seeing 6.9 GHz on LN2. SB couldnt do that.

 

[Nemesis 1]

People expected Ivy to do 5.3Ghz daily on air without issue and if it was an exact die shrink of sandy it would have but IPC and 3D Transistors changed all that

Some People expected 5.2 5.3 ghz on Air . I said 5.2 5.3 on good water would be about it at 1.35 v . maybe it will take more like 1.3 V or less. I will see this weekend . I have NO doubt I will be 24/7 @ 5.2 ghz . That silver base on my cooler should pay off big time with IX.

 

[Nemesis 1]

A 3 to 5 % performance difference is nothing to talk about and certainly doesn't create a 200-300mhz performance gap. Not speaking down on Ivy, just the false information Ivy owners are spreading.

Well its people like you who were BD crazy that are knocking the IB ipc improvements. And your under calling those improvements to top it off . I seen 5-15+% improvements from Ats review. Than the 20-60% increase in IGP . So its a great chip unlike what your tring to say .

 

[TidusZ]

Ivy Bridge is going to boost the sale of automotive sandpaper and indigo xtreme for sure. I'm planning to sand down the IHS with a low grit to really get it nice and thin before polishing.

 

[God Mode]

Does anyone have access to an X-ray so we can see if removing the IHS is viable?

 

[BFG10K]

I’m sticking with my i5 2500K. The small reduction in power consumption is more than offset by the much higher temperatures, which will cause the CPU fan to spin up more.

 

[Smoblikat]

I’m sticking with my i5 2500K. The small reduction in power consumption is more than offset by the much higher temperatures, which will cause the CPU fan to spin up more.

I agree, my 2500K is at 1.05v for 3.7 so im sure im down to around 77w

 

[Smoblikat]

How in the world did you extract that information from my post? Please, I'm dying to know.

By you cherry picking ONE benchmark out of all of them to proove your point that ivy isnt a flop.

 

[BallaTheFeared]

Ivy is a flop now?

 

[Lonbjerg]

Ivy is a flop now?

Apperently everything not AMD "sucks" on the forum and shines in the reviews.
Remember the hype over "piledrivers" romered 10% improvement in IPC...hailed as the second comming of CPU's.
Or remember all those perf/watt arguments.
Now suddenly less power for the same performance is a flop.

It's quite hillaorius...and sad in the same way.

 

[blckgrffn]

Apperently everything not AMD "sucks" on the forum and shines in the reviews.
Remember the hype over "piledrivers" romered 10% improvement in IPC...hailed as the second comming of CPU's.
Or remember all those perf/watt arguments.
Now suddenly less power for the same performance is a flop.

It's quite hillaorius...and sad in the same way.

I salute you for managing to pull that out of this thread. Magnificent!

Excellent points my friend, and I agree. Now I wonder what are the benefits of a 3770K versus a 2700K. Since we're discussing the new chips in the "CPUs and Overclocking" forum in the "Ivy versus Sandy Temp" thread, we must dutifully compare the two architectures in an overclocking and temperature contest. It will be a very close race, and tough to call a winner between the two chips! You make good points: Costs are down, computations are more efficient, power usage is down, there is an increase in performance.

That's the vibe I am getting, man. (pulled from the other IVB thread getting hits right now)

The folks disappointed in IVB are SB owners, so I don't see how that pertains to AMD at all.

The ROI on IVB vs SB is not great. Maybe if you pulled the SB now and sold it, you could come out even after a year or so? By that time Haswell is here, and IVB is in the rear view mirror.

 

[Hulk]

I’m sticking with my i5 2500K. The small reduction in power consumption is more than offset by the much higher temperatures, which will cause the CPU fan to spin up more.

There is so much disinformation regarding Ivy Bridge going around that I feel the need to clarify a few things.

1. Temperature and heat transfer are not the same thing. The joules of heat produced by Ivy are less than Sandy at comparable clocks and volts. If you place a control volume around your system, LESS heat will be produced by Ivy vs. Sandy. Your CPU cooling fan will run SLOWER.

2. Since the heat from the Ivy die has to be transmitted across a smaller surface area the flux is greater, or if it is not, then the temperature of the die will be hotter, which is what we are seeing. There are basically two ways to increase the heat transfer. First is to increase thermal conductivity, i.e. use Cu instead of Al, lapping,etc.. and second is to increase the temperature differential between the high and low temp sources. Water or more exotic cooling methods achieve that.

3. When the heat TRANSFER issue is resolved Ivy actually overclocks better than Sandy as N2 reports around the net are showing.

4. More voltage using standard cooling techniques is a problem with Ivy because it makes the heat transfer issue more prominent.

5. It seems as though Intel designed the chip to work very well at stock speeds but a package redesign might be in order to increase heat transfer for overclockers.

6. As mentioned above I would suspect that lapping, correct application of thermal paste, and a Cu (at least the base) CPU cooler would be helpful.

In summary, Ivy is not a dud. With adequate heat transfer from the substrate it clocks higher than Sandy using less volts, and has a higher IPC than Sandy. Watch as the enthusiast community figures out how to deal with this heat transfer issue.

 

[exar333]

Apperently everything not AMD "sucks" on the forum and shines in the reviews.
Remember the hype over "piledrivers" romered 10% improvement in IPC...hailed as the second comming of CPU's.
Or remember all those perf/watt arguments.
Now suddenly less power for the same performance is a flop.

It's quite hillaorius...and sad in the same way.

Agreed.

5-10% more performance with 10-20% less power usage is EXACTLY what Intel needs to do. Look at the market. IB was more about 22nm and 3D xtors and less about making the CPU OC more and getting tons of extra features. Haswell will tell us if Intel is on the right path or not IMHO.

Let me get this straight:
-'BD': Less IPC and more power usage = 'flop'
-IB: More IPC and less power usage = 'flop'

Does not compute!!

 

[blckgrffn]

There is so much disinformation regarding Ivy Bridge going around that I feel the need to clarify a few things.

1. Temperature and heat transfer are not the same thing. The joules of heat produced by Ivy are less than Sandy at comparable clocks and volts. If you place a control volume around your system, LESS heat will be produced by Ivy vs. Sandy. Your CPU cooling fan will run SLOWER.

About this, please help me understand.

If my board controls the CPU fan in reaction to CPU temperature, how will IVB make the fan run slower? If the CPU is running hotter, then more fan speed will be required, no?

I suppose it comes down to tweaking the reaction of your fans, whether that is in the UEFI or other software, but it seems like a given fan profile is going ramp up more often with IVB vs SB.

 

[Lonbjerg]

Agreed.

5-10% more performance with 10-20% less power usage is EXACTLY what Intel needs to do. Look at the market. IB was more about 22nm and 3D xtors and less about making the CPU OC more and getting tons of extra features. Haswell will tell us if Intel is on the right path or not IMHO.

Let me get this straight:
-'BD': Less IPC and more power usage = 'flop'
-IB: More IPC and less power usage = 'flop'

Does not compute!!

I know...it simply dosn't compute.
Besides people talk like this was a tock..and not a tick(+).

Intel now has a CPU that will scale from ultrabooks to servers...on 22nm...when the rest of the industry is having labourpains with their 28nm.

And they launched a CPU that "Piledriver" won't beat.
At a lesser MSRP than Sandy Brigde.

It even has no cold bug...if you a serious about OC'ing with serious cooling.

Only thing I don't care about in IVB...is the IGP.
So my i7 920 won't get replaced untill 2011 IVB.
Not paying for a useless IGP that I will never use.
(this is my only beef with CPU's these days...the lame IGP's).

Yeah it's a "flop"...but not in reviews...but amongst certain forumposters :sneaky:

I guess you know who I will listen too

 

[blckgrffn]

I know...it simply dosn't compute.
Besides people talk like this was a tock..and not a tick(+).

Intel now has a CPU that will scale from ultrabooks to servers...on 22nm...when the rest of the industry is having labourpains with their 28nm.

And they launched a CPU that "Piledriver" won't beat.
At a lesser MSRP than Sandy Brigde.

Hah, PD coming close to SB in anything outside of highly threaded work loads is going to take a miracle. Forget IVB

Hopefully the platform will be cheaper and Trinity competitive so we continue to see Intel have some competition to keep innovation on pace and prices in check.

If PD/Trinity/Brazos v2 can help do that, in addition to other market forces at work, we all win

Don't be that guy who says we can't all be excited about improvements from AMD though - that's just poor form. Especially if they can do something meaningful on same node with essentially the same architecture.

The development cycles are so long, the hype is just an inevitable part of the event of a new CPU release. We may as well enjoy it (ie, use it as an excuse to participate in this forum.)

 

[Don Karnage]

Hah, PD coming close to SB in anything outside of highly threaded work loads is going to take a miracle. Forget IVB

Hopefully the platform will be cheaper and Trinity competitive so we continue to see Intel have some competition to keep innovation on pace and prices in check.

If PD/Trinity/Brazos v2 can help do that, in addition to other market forces at work, we all win

Don't be that guy who says we can't be excited about improvements from AMD though - that's just poor form. Especially if they can do something meaningful on same node with essentially the same architecture.

Nothing makes me smile more then another Amd Failure

 

[blckgrffn]

Nothing makes me smile more then another Amd Failure

Well its been a good year for you then

 

[Maximilian]

BD is a flop for everyone

IB is a flop for enthusiasts

 

[LOL_Wut_Axel]

Hah, PD coming close to SB in anything outside of highly threaded work loads is going to take a miracle. Forget IVB

Hopefully the platform will be cheaper and Trinity competitive so we continue to see Intel have some competition to keep innovation on pace and prices in check.

If PD/Trinity/Brazos v2 can help do that, in addition to other market forces at work, we all win

Don't be that guy who says we can't all be excited about improvements from AMD though - that's just poor form. Especially if they can do something meaningful on same node with essentially the same architecture.

The development cycles are so long, the hype is just an inevitable part of the event of a new CPU release. We may as well enjoy it (ie, use it as an excuse to participate in this forum.)

Yeah, saying it like it is is "acting on poor form". The truth is that Llano to Trinity will be the same thing as Sandy Bridge to Ivy Bridge when you look at it fundamentally. Intel wasn't able to give an average 50% higher IGP performance going from the HD 3000 to 4000 even with a smaller process node and 3D transistors. AMD is gonna be using the same process node and the same transistors as before, and a CPU that will at last be on par with Stars. And we all know their CPU engineers don't come even close to being as good as Intel's, so if you look at it objectively Trinity isn't that big of a deal.

 

[Hulk]

About this, please help me understand.

If my board controls the CPU fan in reaction to CPU temperature, how will IVB make the fan run slower? If the CPU is running hotter, then more fan speed will be required, no?

I suppose it comes down to tweaking the reaction of your fans, whether that is in the UEFI or other software, but it seems like a given fan profile is going ramp up more often with IVB vs SB.

How about this analogy?

Imagine you have two rooms to cool. The "Sandy" room is big and has lots of windows, it could get really hot but you have two air conditioners in it. It costs quite a bit to keep them running but it keeps the room nice and cool. LOTS of hot air is pumped out of that room by the air conditioners.

The "Ivy" room is smaller than the Ivy room and has less windows so you figure you can just use one air conditioner. It works except when it gets really hot and then the temperature of that room rises a little bit. Still that one air conditioner uses less power and pumps out less heat than the "Sandy" room. And under all but the hottest weather (high overclocking) the Ivy room stays cooler and costs less to keep cool.

The problem is not the amount of heat Ivy is generating, it's getting it out of the die quickly. If you don't move it out then the temps rise. The thing to remember is there is actually less heat energy to move than Sandy but due to the smaller die it's harder to achieve the same heat transfer to the CPU cooler.

As I wrote above I wonder what really good lapping and installation will do for Ivy.

There is nothing wrong with the tri-gate technology or Ivy. It's a beast and hits nearly 7GHz when cooled properly. Intel designed it for the stock speeds thermally, and it works fine. But the big overclocks will require some different cooling techniques.

I'm still thinking Ivy is a good candidate for a hybrid cooling system. Perhaps a small sub zero cooler just to create a high temperature differential from the die to the sub zero cooler, then a normal air cooler to cool the sub zero cooler.

 

[blackened23]

It isnt that hot. I mean it is hotter than Sandy Bridge when we compare clock to clock or voltage to voltage but overall its fine. Its fine even on air cooling. See this, 4.64 GHz and load temps are 75 with Venomous X: http://www.maximum-tech.net/intel-core-i7-3770k-3-5-ghz-ivy-bridge-review-12192/

though IVB seems to scale really well with cold. We are already seeing 6.9 GHz on LN2. SB couldnt do that.

I don't really see it that way. To get acceptable temps you need to get an oc at around 1.1V-1.15V, and most reviewers cannot get past 4.5ghz without 1.25V or more. At 1.1V it is already a lot warmer than sandy bridge, and once you get up to 1.25V-1.3V it starts hitting 85C to 90C.

Its a wonderful CPU for water coolers but for air coolers? Disappointing...just IMO. Not sure why BD is being brought up, we all knows its junk, intel is really competing with themselves as far as desktop CPU's are concerned...and Ivy bridge is the most disappointing tick that I can remember from recent years. I guess its okay at stock setting, but I was really hoping for better air overclocking headroom. Every new node/die shrink in the past 5 years has brought better overclocking headroom until now, thats my main reason for being really disappointed.

 

[DarkRogue]

Hulk, I agree that IVB should generate less heat, and the main problem is with the transfer of that heat away from the CPU.

However, I believe blckgrffn has a point with those fans.
If you have your system control the fans, they are typically tied to the temperature of your CPU cores.

Since IVB has trouble transferring the heat away as quickly/efficiently, even though it's producing less of it, it builds up and the cores start getting warmer. As a response, the system would ramp up the fan speed in an attempt to cool it down, but it's not as effective simply because the heat isn't making it to the heatsink very efficiently.