The Core i7-4770K Review: Haswell Is Faster; Desktop Enthusiasts Yawn

[Sakkura]

[...] Android [...] is designed with touch input in mind. [...] Android is currently far from being mouse-friendly

"Most people" are not attached to mouse input. If they can ban it, they will. Anyways, basic mouse interaction (for basic word processing, spreadsheet, aso) is already supported. The next "desktop" will be a semi horizontal tablet, laying in a flatbed dock surrounded by a small keyboard (lower border), a stylus, and, maybe, a wireless mouse.
Power users (everybody but "most people") will use "true desktop" OSes that are still evolving beside the tablet craze.

It's not the mouse that is central, it's the keyboard. There are certain tasks that just call for a keyboard (and not some half-arsed useless imitation). And if you're using a keyboard, the logical complement is a mouse or equivalent pointing device. Now, a lot of people don't need a keyboard since they mainly receive information from the device and rarely input significant amounts of information. But that doesn't apply to everyone.

 

[InvalidError]

"Most people" are not attached to mouse input. If they can ban it, they will. Anyways, basic mouse interaction (for basic word processing, spreadsheet, aso) is already supported. The next "desktop" will be a semi horizontal tablet, laying in a flatbed dock surrounded by a small keyboard (lower border), a stylus, and, maybe, a wireless mouse.

I suggest you try using a tablet with a keyboard while the tablet is laying flat. You will likely get a sore neck in record time. Once you raise the tablet's height to a more comfortable eye-level height to avoid neck-breaking postures, lifting arms to interact with the touch-screen will give you sore arms over time and make you wish for a mouse.

Not to mention that touch-screen fares pretty badly at precision and speed vs mouse for everyday trivial things like selecting text.


While Android may have work-arounds for mouse-less inputs, they cannot match the speed and precision. Just like how on-screen keyboards cannot match real keyboard's speed and precision.

 

[fiduce]

I suggest you try using a tablet with a keyboard while the tablet is laying flat.

Semi-horizontal is not "laying flat".

 

[InvalidError]

Semi-horizontal is not "laying flat".

There is no angle that will be simultaneously ergonomically correct for both the keyboard and display for extended use on a laptop or tablet with keyboard dock: putting the keyboard at a correct height brings the display too low and causes neck strain while putting the laptop high enough to avoid neck strain brings it too high for arms and cause extra strain there.

Only way to make a laptop/docked tablet ergonomically optimal is to either use external display or external keyboard.

 

[fiduce]

There is no angle that will be simultaneously ergonomically correct for both the keyboard and display for extended use on a laptop or tablet with keyboard dock

So guess how we've been using paperback books and paper + pens for ages ? Paper was simply laying on the desk !
How do you read your paper mail ? like a medical radiography (sticked to a vertical lightbox) ?
If you lay a tablet 45 degrees down, in front of you, you won't get armed and you will be able to touch effortless either a small keyboard in front of it, or its screen. If you rather use a mouse, then connect a wireless mouse (click = touch, drag = swipe)

 

[InvalidError]

So guess how we've been using paperback books and paper + pens for ages ? Paper was simply laying on the desk !

How do you read your paper mail ? like a medical radiography (sticked to a vertical lightbox) ?

You never got sore neck/back from crouching over a table to write with pen and paper? Just because that's how most people do it does not make it ergonomically correct. The closest thing to an ergonomically correct way of using pen and paper is to use an angled drawing table.

As for how I read stuff, that's usually by holding the sheets or book at a comfortable angle near eye-level unless the book is too heavy to hold long enough for me to finish reading whatever I need to read from it.

 

[slomo4sho]

lol.

With that train of thought, buying AMD chips supports AMD's "business decision" to be over two generations behind Intel CPU performance wise.

By buying into Ivy Bridge instead of Haswell, he is voting with his cash on Ivy Bridge being the better value at this point in time. If enough people choose to do the same (however unlikely that is), Intel would eventually have to admit people did not support Intel's business decisions - somewhat like how end-user backlash ended up forcing Microsoft to admit more people cared about the Start Menu than they thought they did.

Of course, avoiding Haswell will become increasingly difficult once Intel starts discontinuing Ivy Bridge models 6-8 months from now.

Intel doesn't drop prices on its older generation processors and revenue is revenue regardless of which processor you end up with. The quarterly reports will still show success if you purchase Ivy over Haswell.

A monopoly (yes, Intel is currently a monopoly) will not learn anything until it starts to lose substantial market share.

 

[InvalidError]

Intel doesn't drop prices on its older generation processors and revenue is revenue regardless of which processor you end up with. The quarterly reports will still show success if you purchase Ivy over Haswell.

Orderbooks and stock-keeping does tell Intel which products people want and which products people don't want. If Haswell turned out to be so bad that every company and remotely educated buyer would actively avoid it, Intel would have to reconsider their plans sooner rather than later due to failing to shift Haswell stock out.

 

[slomo4sho]

Orderbooks and stock-keeping does tell Intel which products people want and which products people don't want. If Haswell turned out to be so bad that every company and remotely educated buyer would actively avoid it, Intel would have to reconsider their plans sooner rather than later due to failing to shift Haswell stock out.

Performance and price being equal on both chips, is there any reason to think that the general public wouldn't simply opt for haswell because it is "new"? It's not like Intel doesn't have an Ad campaigns out glorifying the new chips.

 

[InvalidError]

Performance and price being equal on both chips, is there any reason to think that the general public wouldn't simply opt for haswell because it is "new"?

The original argument behind this thread fork was about some people disagreeing with the direction Intel has taken with Haswell. All I was saying is that if Intel's new direction really was that bad, people would ignore Haswell and buy something else, be it from Intel's other product lines or AMD.

But reality is that Haswell is indeed generally better than IB for non-enthusiasts so there is no reason for normal PC buyers to bother being picky.

 

[redeemer]

[citation][nom]redeemer[/nom]10% increase over Sandy Bridge is a myth, too much heat too little gains Sandy Bridge is still superior i5 and i7 hitting 5.0 Ghz with air cooling something that is impossible on Ivy Bridge.[/citation]

Myth? Every bench shows higher performance per clock when comparing Sandy vs. Ivy bridge. What kind of myth are you talking about? It has native USB 3 and better power consumption and you complain about maximum overclocks? Ivy get's more done with a lower frequency and my i7-3770K is running 4.5Ghz on air. It idles at 32 and hits low 60s with all 4 running 4.5.

Ivy Bridge gets more done how exactly adding 1-2fps to you gaming experience over Sandy Bridge. What about the high temps? I get over 10.0 on my cinebench score an 3770k OC to 4.8Ghz scores lower, so much for the 10% increase right?? Again all chips are different maybe yours requires less voltage, but generally like again all reviews as you pointed out also show the Ivy temps going nuts after 1.28v? So Sandy can clocks higher and stay cooler this is important for a K unlocked processor. Overclocking is not about benchmarks only its about hitting magical numbers!

 

[ojas]

[citation][nom]redeemer[/nom]10% increase over Sandy Bridge is a myth, too much heat too little gains Sandy Bridge is still superior i5 and i7 hitting 5.0 Ghz with air cooling something that is impossible on Ivy Bridge.[/citation]

Myth? Every bench shows higher performance per clock when comparing Sandy vs. Ivy bridge. What kind of myth are you talking about? It has native USB 3 and better power consumption and you complain about maximum overclocks? Ivy get's more done with a lower frequency and my i7-3770K is running 4.5Ghz on air. It idles at 32 and hits low 60s with all 4 running 4.5.

Ivy Bridge gets more done how exactly adding 1-2fps to you gaming experience over Sandy Bridge.

The world doesn't run on games, son. You need computers for other things too.

 

[samal90]

The reality of computing today is that 90% of people just use their computer to browse the net, type documents and chat on skype..maybe watch some videos and play the occasional casual game. The processing power of 4 year old CPUs are more than enough to do that. The rest is just marketing to put it in people's minds that they always need the next best thing for whatever reason..."your webpage will load 1 sec faster..OMG...imagine the time you'll save" LOL. The war that is coming will be on integrated GPU and AMD has an advance in that position. I respect intel for their incredible processing power CPUs but unfortunately they lack integrated GPU strength...at an AFFORDABLE price. AMD on the other hand is the contrary...lackluster CPU but good GPU integration. At the end of the day, it all depends where your priorities lie. Personally, I would like to have an thin laptop that is able to play games decently without the need for a dedicated GPU. Therefore, I will wait for Kaveri at the end of the year to see the competition. If its underwhelming, i'll buy a haswell ultrabook...End of story.

 

[InvalidError]

I respect intel for their incredible processing power CPUs but unfortunately they lack integrated GPU strength...at an AFFORDABLE price.

I can't help thinking they missed the mark by ~$200 on GT3e if they really want ultrabook sales based on it to take off.

There is always hope that they will fix that with Broadwell where they will have the transistor budget to make something even better than GT3 (non-e) standard.

 

[Joyce Ann Tamayo]

I can't see people buying Intel products here in PH. As for Asian people, more investor use AMD apu's on there internet shop which is reliable w/o integrated GPU but more RAM for online game.. year after year it drops..

 

[ojas]

I respect intel for their incredible processing power CPUs but unfortunately they lack integrated GPU strength...at an AFFORDABLE price.

I can't help thinking they missed the mark by ~$200 on GT3e if they really want ultrabook sales based on it to take off.

There is always hope that they will fix that with Broadwell where they will have the transistor budget to make something even better than GT3 (non-e) standard.

Ultrabooks won't get GT3e...only GT3.

 

[InvalidError]

Ultrabooks won't get GT3e...only GT3.

Depends on who you ask.

For people who view ultrabooks as nothing more than a form factor, there is no intrinsic problem with cramming a 55W part in there. Sure, it won't meet Intel's whole-day computing definition but people who only care about the form factor won't mind that.

 

[Wisecracker]

Ultrabooks won't get GT3e...only GT3.

Depends on who you ask.

For people who view ultrabooks as nothing more than a form factor, there is no intrinsic problem with cramming a 55W part in there. Sure, it won't meet Intel's whole-day computing definition but people who only care about the form factor won't mind that.

I'm not the expert on the Intel graphics engine but I think it's going to work like this: GT3e is a 65w desktop part, and the Ultra-line portables get GT3u which is clocked 200-300MHz slower but consumes half the wattage. There are some Intel slides floating around about this.

Hard to say where GT3u will ultimately shake out, but Kaveri will likely smack it around as AMD has been uncharacteristically silent (other than the mobile 'Mars XTX' which likely is the GCN core for the new APU).

 

[InvalidError]

I'm not the expert on the Intel graphics engine but I think it's going to work like this: GT3e is a 65w desktop part, and the Ultra-line portables get GT3u which is clocked 200-300MHz slower but consumes half the wattage.

GT3e is also available on mobile parts:
http://ark.intel.com/products/76085
47W max TDP at 3.6GHz.

A single ~47W chip is much simpler to power, cool and design around than a 35W CPU + 40W discrete GPU and those are exactly the arguments Intel uses to justify the $100-200 premium for GT3e over equivalent or better discrete graphics.

What I would like to see is AMD usurp most of the mid-range pseudo-ultrabook market with laptops in the $500-800 range that put GT3e to shame at least GPU-wise. This would force Intel to admit that $450-650 is not a sane price range for GT3e chips.

 

[ingtar33]

I'm not the expert on the Intel graphics engine but I think it's going to work like this: GT3e is a 65w desktop part, and the Ultra-line portables get GT3u which is clocked 200-300MHz slower but consumes half the wattage.

GT3e is also available on mobile parts:
http://ark.intel.com/products/76085
47W max TDP at 3.6GHz.

A single ~47W chip is much simpler to power, cool and design around than a 35W CPU + 40W discrete GPU and those are exactly the arguments Intel uses to justify the $100-200 premium for GT3e over equivalent or better discrete graphics.

What I would like to see is AMD usurp most of the mid-range pseudo-ultrabook market with laptops in the $500-800 range that put GT3e to shame at least GPU-wise. This would force Intel to admit that $450-650 is not a sane price range for GT3e chips.

the thing is we're not seeing iris pro in anything. I thought for certain at it's insane price premium it would be in the new mac book pros, but it's not even going into those. Too pricey for apple? didn't think i'd see the day.

That makes this an engineering trick, not a finished product. This is a product priced too high for the market. Sorta like one of those quantum computers that go for 10mil... no one really needs one, there is no real market for it. it's just an engineering trick.

Until the iris pro hits market in any relevant way this is just an intel bench-marking trick... nothing more; and if Kaveri is the ballpark of iris pro performance at typical amd low prices it could be an opportunity for AMD to shoehorn itself into another market; and maybe force intel to actually compete for a change.

 

[Wisecracker]

I'm not the expert on the Intel graphics engine but I think it's going to work like this: GT3e is a 65w desktop part, and the Ultra-line portables get GT3u which is clocked 200-300MHz slower but consumes half the wattage.

GT3e is also available on mobile parts:
http://ark.intel.com/products/76085
47W max TDP at 3.6GHz.

A single ~47W chip is much simpler to power, cool and design around than a 35W CPU + 40W discrete GPU and those are exactly the arguments Intel uses to justify the $100-200 premium for GT3e over equivalent or better discrete graphics.

What I would like to see is AMD usurp most of the mid-range pseudo-ultrabook market with laptops in the $500-800 range that put GT3e to shame at least GPU-wise. This would force Intel to admit that $450-650 is not a sane price range for GT3e chips.

Sorry for not being a bit more clear ...

The masks used for the Mars XTX mobile discreet likely cross-over to the Kaveri APU SIMD engine array. I was not referring to the mobile discreet version. While I don't know how exacting the selection is, TSMC/AMD has had several years perfecting the Cape Verdi GCN cores. At introduction of the desktop discreet, the best folks were hoping for was a level of performance similar to a cut-down HD7750 for the Kaveri APU. Nobody is talking, but the anticipated level of performance has grown beyond that (based upon some cryptic AMD slides).

And, no offense, but holding up the i7-4950HQ (CPU World projected tray cost: $657) is kinda wacky. While hats are off to Intel for boosting their graphics performance, no one is going to purchase that chip for HD5200-level performance.

 

[ojas]

@Invalid: The ultrabook spec only allows 15, 17 and 28w parts. 47w is too much for an ultrabook, at least without seriously re-thinking cooling.

Present day ultrabooks can barely handle 17w.

What i'm saying is, notebooks that'll carry the ultrabook tag won't get anything more than GT3. I'm beyond certain about this.

@Wisecracker: I read on Seeking Alpha that bulk orders for a SKU cost half as much per chip than the 1K price...so it's a $330 chip for OEMs...though for boutique builders, $657.

 

[InvalidError]

@Invalid: The ultrabook spec only allows 15, 17 and 28w parts. 47w is too much for an ultrabook, at least without seriously re-thinking cooling.

Who cares about Intel's official ultrabook specs? That has never stopped OEMs from doing their own thing without Intel's marketing subsidies. If an OEM wants to cram a 45W part in their own ultrabook-like device, Intel isn't going to stop them.

Higher power parts have been crammed into smaller than intended form factors in the past and OEM will certainly do it again if they see a large enough market opportunity to forgo Intel's blessings.

 

[ubercake]

Why do we have holes in the motherboard for 1150 cooling solutions? Why not just let us screw the cooling solution down to the board like AMD or LGA 2011?

Why doesn't Intel see built-in motherboard mounts for a cooling solution as an improvement and just implement into their base spec?

Having to pop a cooling solution through your motherboard puts unnecessary stress on the motherboard. Having to use a backplate is also an incovenience, since many cases don't have cutouts requiring complete disassembly of the system to mount a new cooler. Some cases that do have cutouts have cutouts that don't always line up with the backside of the processor mount also requiring disassembly in some cases.

This is very basic. We see advancements with everything and having the cooler mounts on the motherboards seems to be consistently overlooked by Intel. Next thing you know, we'll be using chip pullers to update our RAM and VRAM memory again!?

 

[InvalidError]

Why doesn't Intel see built-in motherboard mounts for a cooling solution as an improvement and just implement into their base spec?

There were a few iterations of that in the past, starting with socket-based hooks directly on the 486's socket until motherboard-premounted anchors on the P4/s478. s478's cam-lock is my favourite HSF attachment method to date: easy to put on, fairly easy to take off, lots of "spring" in the design that make the cams' clamp-down effect fairly obvious and enough resistance from the cams while setting them to give a satisfying sense that the contraption is not going to go anywhere.

I really hate push-pins for how little feedback they give about how much force is actually being applied on the CPU so the only way to know for sure whether or not they are reasonably well set is to have a look at idle and load temperatures.