• Competitor rules

    Please remember that any mention of competitors, hinting at competitors or offering to provide details of competitors will result in an account suspension. The full rules can be found under the 'Terms and Rules' link in the bottom right corner of your screen. Just don't mention competitors in any way, shape or form and you'll be OK.

Haswell refresh price list leaked

Man of Honour
Joined
21 May 2012
Posts
31,947
Location
Dalek flagship
Ten upcoming parts priced

Haswell refresh is just weeks away and it appears that most of them will end up with a 100MHz clock bump. Don't expect much in the way of new features and tweaks - this is still essentially a one-year-old architecture and Intel doesn't do miracles.

According to CPU World, which dug up a Haswell refresh price list on ShopBLT, almost all of the chips will get a 100MHz base clock bump and boost clocks will be increased by 100MHz, too.

The prices aren't surprising, either.

The entry level Celeron G1840 is priced at $47.50, followed by the G1850 at $59. Two Pentiums are in the mix, the G3240 and G3440, priced at $70 and $90 respectively.

Three new Core i3 SKUs are coming, priced between $132 and $166. The Core i5 4590 and Core i 5 4690 will cost $213 and $235 respectively. The lone Core i7 part is the i7 4790, priced at $326.50.

The chips should go on sale sometime in Q2, but we are not entirely sure when.

Preorder prices of Haswell-R desktop CPUs

http://www.fudzilla.com/home/item/34095-haswell-refresh-price-list-leaked
 
Dat ~3% improvement.

Waste of time and effort tbh, the only way this becomes worthwhile/news worthy IMO is if the silicon has notably better overclocking and/or soldered IHS.
 
Any chance of the top I7****K's supporting all the virtualisation instructions/IO this time or will it be disabled on the 'K' versions again.
 
Theyre not bad chips, but like ib the terrible paste job Intel does really doesn't help on what is supposed to be an overclocking chip.
 
Is there anyone actually impressed at all by the Haswell family of CPUs?

Haswell bar its flaws is impressive, some of the FPU gains are quite substantial, no where near as high as you'd want for over 2 years of gain, but at least it's consistent and there.

I guess the problem isn't with Haswell, but what's available, really I'd want a 6 hex core (Forget HT for now) at around 300 quid that I can just slot in a 1150 board, that'd make Haswell seem more impressive, that's not because of Haswell, but rather it's 6 modern Intel cores at an upper mainstream price.
 
Die size......

I don't buy it, and if they throw more into the IGP (GT3 for example) as they intend to for Broadwell, the die size will increase a fair whack, making it moot if it really is the issue.

Or a hexcore mainstream part, and this die size problem is no more.

They could throw GT3 in the refresh with solder, that'd make the refresh worthwhile.
 
I don't buy it, and if they throw more into the IGP (GT3 for example) as they intend to for Broadwell, the die size will increase a fair whack, making it moot if it really is the issue.

Or a hexcore mainstream part, and this die size problem is no more.

They could throw GT3 in the refresh with solder, that'd make the refresh worthwhile.

It will not happen.... As the dies get smaller because of a small process the problem will get worse.

The dies are so small that putting so much heat through a small amount of solder causes the solder to eventually break down and crack, reducing performance in the process.

As I've said above, the socket 2011 uses much larger die sizes where this solder fatigue does not occur. But as these dies also get smaller in the future they will also have to move to TIM.

There's a big white paper on the net that Intel released about the problem, includes magnified shots of solder being physically cracked from being heated and re-heated in a such a small concentrated area.
 
The 4820's not exactly more expensive than the 4770K or the 3770K before it, bring those features to the mainstream, progression and all that.
 
My suggestion for paste then, intel should start using liquid pro/ultra and a bit less black sealant on the ihs. Due to the inconsistency in how this is applied leads to the wide temp variations on hw chips. On some the ihs is closer to the die, on others it's further apart.

Ive no idea what sort of paste intel use, but it cant be great stuff either. From pics/videos you see it doesnt look like a high end product. On a delidded chip the temp drops from using liquid pro/ultra can be very substantial.
 
Back
Top Bottom