Become a MacRumors Supporter for $50/year with no ads, ability to filter front page stories, and private forums.
After all, the HD 4000 is almost as fast as the 9600M GT Apple used in the 2009 MacBook Pros, which was a mid-class graphic card back then. Not too bad for just three years I'd say. Give Intel another three years and we can get GT 650M performance from a onboard graphics chip, which is absolutely awesome!

I know I am quoting an old post but that will take too long. Intel needs to step up its game further than the 650M in three years in my view.
 
I know I am quoting an old post but that will take too long. Intel needs to step up its game further than the 650M in three years in my view.



According to the link I posted above the Intel HD 4000 has similar performance to the 650M (134 vs. 126).
 
According to the link I posted above the Intel HD 4000 has similar performance to the 650M (134 vs. 126).

:)))) hope u do not believe that;)) hd4000 is downgrade in comparison to ati 6330m, which is so inferior to ati 6770m and this one is inferior to nvidia 650m... Hd4000 is crap
 
In fact, I'm still stunned how fast the 9400M GT in my MacBook is (which complements the 9600M GT), it's just that we loose any possibility to play games -- although as far as I've seen SC2 and Diablo 3 seem to run just fine on low.

Even last year's HD3000 is about as fast as the older discrete graphics (9600M GT). It's all relative.

My 2011 server with the integrated graphics plays some games just fine (Source engine primarily) albeit not with all the graphics settings cranked up.

There's nothing wrong with integrated graphics for normal use and light gaming. If your primary use is gaming, don't buy a mini.... if I really want to game I'll hook my PS3 into the HDMI input of my new 27" Dell monitor. :)
 
:)))) hope u do not believe that;)) hd4000 is downgrade in comparison to ati 6330m, which is so inferior to ati 6770m and this one is inferior to nvidia 650m... Hd4000 is crap


I haven't used all of those GPUs so can't comment. However from the data it tells a different story. Do you have any numbers to show otherwise?
 
Good Luck...

I haven't used all of those GPUs so can't comment. However from the data it tells a different story. Do you have any numbers to show otherwise?

Judging from his post I wouldn't hold your breath waiting for an intelligent response to your request for some data that supports his assertion that the HD4000 is "crap"...

Take a look at http://www.notebookcheck.net/Intel-HD-Graphics-4000.69168.0.html.

And here is another excellent website: http://www.videocardbenchmark.net

Barefeats.com has myriad tests comparing the performance of the HD4000 with other current Macs.

Using those sources you should be able to come to your own conclusion about how well the Intel HD4000 will meet your computing needs.
 
Indeed, the HD4000 is crap for a 2012 2.6 quad-core machine. In my own tests, my 2011 dual-core 2.5 Mini with ATI-6630m had a higher Cinebench (23.75 vs 22.25) that the 2012 2.6 quad! I returned the 2012 Mini, and kept the 2011.
 
My 2012 Mini is absolutely blinding, and I have no evidence of any shortcomings with the GPU.

I'm driving two 20" Apple Cinema Displays, doing reasonably heavy Photoshop and Illustrator CS6. I use a 3D design package called "Sweet Home 3D" and Microspot 3D Toolbox. I play "Armagetron", a light-cycle game and "Oolite", a space-flight trading & combat game. Both are OpenGL.
Armagetron claims to give me over 300 fps, and I'm getting over 50 in Oolite. The 3D design apps work perfectly well, with instant modelling of the 3D objects. OK, it's not Pixar...

Maybe there are tasks that the Intel 4000 isn't good enough for, but they are few and at the 3D pro end of the spectrum -- for which a mobile GPU probably isn't good enough either.

But the sheer power of the CPU makes this computer absolutely fly in any real-world use, GPU benchmarks be damned.

BTW: I recommend all the above apps, which are all free.
 
What may be "crap" to one person may be fine for another; blanket statements serve no one who is trying to determine whether a particular GPU will meet their specific needs.

Frankly, many, if not most Mac Mini users will not even notice the GPU performance of their Mac. The Mac Mini is not intended to be used as a high-end gaming computer. It may not be the best choice for those who intend to use Final Cut Pro, Photoshop or any other program that benefits from a powerful discrete GPU. However, the Mini may be just fine depending on what you intend to do with those programs. Photoshop performance is fine on my 2012 2.3GHz Mini but the kind of image editing that I do is relatively modest and I am not working in a production environment.

If a person is a member of the subset that is impacted by the performance of an integrated GPU then he/she should select another computer that meets their needs...
 
Last edited:
So I just got my 2012 Mini today (2.3 i7, 16gb, Fusion). Everything is up-and-running. I'm using my LG LED from HDMI -> MiniDisplayPort. The only issue I have noticed today is occasionally the menu bar disappears until you click somewhere else on the screen.

Has anyone else experienced this?
 
Judging from his post I wouldn't hold your breath waiting for an intelligent response to your request for some data that supports his assertion that the HD4000 is "crap"...

Take a look at http://www.notebookcheck.net/Intel-HD-Graphics-4000.69168.0.html.

And here is another excellent website: http://www.videocardbenchmark.net

Barefeats.com has myriad tests comparing the performance of the HD4000 with other current Macs.

Using those sources you should be able to come to your own conclusion about how well the Intel HD4000 will meet your computing needs.

Think twice before you insult anyone...

Have personal experience with hd3000, hd4000, currently using ATI 6750... just without any numbers, I can simply post fact: hd4000<<<<<<<6750M<650m...

if you want numbers, notebookcheck gives you all numbers - ati 6330m stronger than hd4000, others are so clearly different level GPUs that it is not even necessary to comment..

whoever even in dream compares hd4000 with 650m has no idea what he is talking about...

Edit: oh and I forgot, with those ATI/NVIDIA, you wont have to think about workarounds to some great hd4000 isueas like blackouts, whitescreens, etc...
 
Indeed, the HD4000 is crap for a 2012 2.6 quad-core machine. In my own tests, my 2011 dual-core 2.5 Mini with ATI-6630m had a higher Cinebench (23.75 vs 22.25) that the 2012 2.6 quad! I returned the 2012 Mini, and kept the 2011.

So what you are saying is that the HD4000 is only slightly behind the ATI 6630M? Yep that's what all the benchmarks have proven. Not sure what your point is? So you can either have twice the CPU power and give up 6% in the GPU, or give up half the CPU power and gain 6% in the GPU.... Neither GPU is going to rock the gaming world are they? :)

I guess I'd take the faster processor since I'm more likely to use the CPU power than the GPU 90% of the time
 
So what you are saying is that the HD4000 is only slightly behind the ATI 6630M? Yep that's what all the benchmarks have proven. Not sure what your point is? So you can either have twice the CPU power and give up 6% in the GPU, or give up half the CPU power and gain 6% in the GPU.... Neither GPU is going to rock the gaming world are they? :)

I guess I'd take the faster processor since I'm more likely to use the CPU power than the GPU 90% of the time

black-preacher-meme-generator-preach-it-brother-84c7bf.jpg


I support this post.
 
So what you are saying is that the HD4000 is only slightly behind the ATI 6630M? Yep that's what all the benchmarks have proven. Not sure what your point is? So you can either have twice the CPU power and give up 6% in the GPU, or give up half the CPU power and gain 6% in the GPU.... Neither GPU is going to rock the gaming world are they? :)

I guess I'd take the faster processor since I'm more likely to use the CPU power than the GPU 90% of the time

50/50.

As I left the Windows world (I had quite a "server machine" as my friends called it) and got into Timmy's inferno, I can say that I'm quite happy with the 6630 running Dolphin, which I couldn't do with the HD4000.

Hopefully Haswell's "HD5000" will be on par with the 6630M or even more.

In the end it depends on what you plan to use it for...
Was I NOT playing any console emulators at all, I'd pick the new mac mini over the 2011, no questions asked.
 
You Missed Entirely My Point... Which Proves The Sentence You Find Insulting

Think twice before you insult anyone...

In the end comparing the numbers means nothing...

It's all about whether the GPU will meet the needs of a particular computer user PERIOD

Anything else is a meaningless debate, which is what this thread is devolving into... :rolleyes: (Although I think that Paul provided some much-needed insight into what the numbers actually mean when comparing those GPUs.)
 
Last edited:
Judging from his post I wouldn't hold your breath waiting for an intelligent response to your request

If this is not an insult... whatever, this leads to nothing.. all I wanted to say is that HD4000 simply can not be compared to those gpus above... it is weaker gpu than ati 6330m - not just by benchmarks, but even by every day usage experience... hd4000 is not powerfull enough to give you smooth experience with os x using 27inch high res monitor... and that is why i can call intel4000 a crap.

Done with you. have a nice day.
 
According to the link I posted above the Intel HD 4000 has similar performance to the 650M (134 vs. 126).

There is absolutely no way that is possible and I know others have proved it before me though I am just getting to the post now.
 
So what you are saying is that the HD4000 is only slightly behind the ATI 6630M? Yep that's what all the benchmarks have proven. Not sure what your point is? So you can either have twice the CPU power and give up 6% in the GPU, or give up half the CPU power and gain 6% in the GPU.... Neither GPU is going to rock the gaming world are they? :)

I guess I'd take the faster processor since I'm more likely to use the CPU power than the GPU 90% of the time

Not so fast my friend, you didn't read close enough :)... a dual-core 2011 Mini 2.5/Radeon-6630m/1333MHz-bus/256GPU-RAM beat a quad-core 2012 2.6/HD4000/1600Mhz-bus/768GPU-vRAM in a Cinebench Open GL test. So basically the Radeon-6630m is giving up a delta of two cpu cores, 0.1 Mhz cpu speed, 267Mhz bus speed, and 512MB-vRAM and is still beating the HD4000 by 6%! That hardly means the 6630m is only 6% faster, it's only 6% faster on the 2011 platform in the Cinebench OpenGL test.

In fact, notebookcheck.net shows the 6630m is 63% faster than the HD4000 in the 3Dmark11 test (999.8 vs 613.6) which is probably a better indication of their relative speeds on similar platforms.
6630m>
http://www.notebookcheck.net/AMD-Radeon-HD-6630M.43963.0.html
HD4000>
http://www.notebookcheck.net/Intel-HD-Graphics-4000.69168.0.html

The 2012 Mini 2.6 is a beautiful machine, but its Achilles heel is the pitiful HD4000, which is way underpowered for this class of cpu. Obviously, Apple did this by design so as to not undercut iMac and MB sales! Too bad!
 
Not so fast my friend, you didn't read close enough :)... a dual-core 2011 Mini 2.5/Radeon-6630m/1333MHz-bus/256GPU-RAM beat a quad-core 2012 2.6/HD4000/1600Mhz-bus/768GPU-vRAM in a Cinebench Open GL test. So basically the Radeon-6630m is giving up a delta of two cpu cores, 0.1 Mhz cpu speed, 267Mhz bus speed, and 512MB-vRAM and is still beating the HD4000 by 6%! That hardly means the 6630m is only 6% faster, it's only 6% faster on the 2011 platform in the Cinebench OpenGL test.

In fact, notebookcheck.net shows the 6630m is 63% faster than the HD4000 in the 3Dmark11 test (999.8 vs 613.6) which is probably a better indication of their relative speeds on similar platforms.
6630m>
http://www.notebookcheck.net/AMD-Radeon-HD-6630M.43963.0.html
HD4000>
http://www.notebookcheck.net/Intel-HD-Graphics-4000.69168.0.html

The 2012 Mini 2.6 is a beautiful machine, but its Achilles heel is the pitiful HD4000, which is way underpowered for this class of cpu. Obviously, Apple did this by design so as to not undercut iMac and MB sales! Too bad!

Exactly! Giving mini decent gpu it would tear almost any imac config;) its pitty, mini could have been best desktop computer ever made by apple , but with hd it is just another in a row of compromise computers made by apple - there is always a catch with them:(
 
Not so fast my friend, you didn't read close enough :)... a dual-core 2011 Mini 2.5/Radeon-6630m/1333MHz-bus/256GPU-RAM beat a quad-core 2012 2.6/HD4000/1600Mhz-bus/768GPU-vRAM in a Cinebench Open GL test. So basically the Radeon-6630m is giving up a delta of two cpu cores, 0.1 Mhz cpu speed, 267Mhz bus speed, and 512MB-vRAM and is still beating the HD4000 by 6%! That hardly means the 6630m is only 6% faster, it's only 6% faster on the 2011 platform in the Cinebench OpenGL test.

No I read completely what you posted. OpenGL will basically only test the GPU. The rest of the system DOES NOT MATTER. The only thing the CPU will do is feed data to the GPU, but that is minimal since the whole point of the OpenGL benchmark is to test the GPU.

Here's a link directly to the Overview of the tests within Cinebench that confirms that: http://www.maxon.net/products/cinebench/overview.html

The OpenGL benchmark is to test the GPU. Giving up two cores and any cpu speed, really means nothing in this test. Further, as far as giving up VRAM, that is true, but since the 6630m has dedicated VRAM that runs at much faster speeds than the system RAM, I'd call this a push at best.
 
Is that the case? I have a Mac Pro 1,1 with the GeForce 8800 GT...would the new Mini be faster or slower than what I already have?

The gforce graphics card has a 768 bechmark and the hd4000 has got a 479 benchmark but the mac pro 1,1 just gives the gforce 256mb gram. The graphic card is in theory still faster than a hd4000.
For the rest the overall bechmark of the Mac pro 1,1 is lower than the 2011 and 2012 mac mini's. The 2012 i7 2.6 has got a 11367 benchmark, and the macpro 1,1 3.0 has got a 5788 benchmark.

So the mac mini 2012 is up to 2 times faster than the mac pro 1,1.
 
So... help me understand... I see SO much complaining about this... What is the negative impact to "YOU" of HD4000 graphics. What exactly will be the cost? A little slower screen redraws? A little slower filter application in photoshop? ... Some monitors not work at all? Is it just all about games??


People want high end performance for low end price.

HD4000 is good enough for 99% of things including casual gaming on titles a few years old.

And yes, most of it is about games.
 
No I read completely what you posted. OpenGL will basically only test the GPU. The rest of the system DOES NOT MATTER. The only thing the CPU will do is feed data to the GPU, but that is minimal since the whole point of the OpenGL benchmark is to test the GPU.

The OpenGL benchmark is to test the GPU. Giving up two cores and any cpu speed, really means nothing in this test. Further, as far as giving up VRAM, that is true, but since the 6630m has dedicated VRAM that runs at much faster speeds than the system RAM, I'd call this a push at best.

You've got to be kidding-- the cpu speed and memory does matter in the Cinebench OpenGL score, otherwise, e.g. the 2012 Mini 2.3 and 2.6 would have the same score! Go run the test and then compare your score with the population of testers. You will see a graph that shows discrete levels of performance for each cpu version of the 2012 (2.3/2.6), AND within each level, the score improves with more memory!

The 6630 is vastly superior because it is using DDR-5 memory and basically processes each video frame without going out to the bus, even with 1/3 the memory.
 
People want high end performance for low end price.

HD4000 is good enough for 99% of things including casual gaming on titles a few years old.

And yes, most of it is about games.

So you say mac mini is a low end prices computer:D It is not all about gaming, intel hd 4000 is not capable to run osx with its system animation smoothly without lags on apple tb displays

----------

You've got to be kidding-- the cpu speed and memory does matter in the Cinebench OpenGL score, otherwise, e.g. the 2012 Mini 2.3 and 2.6 would have the same score! Go run the test and then compare your score with the population of testers. You will see a graph that shows discrete levels of performance for each cpu version of the 2012 (2.3/2.6), AND within each level, the score improves with more memory!

The 6630 is vastly superior because it is using DDR-5 memory and basically processes each video frame without going out to the bus, even with 1/3 the memory.

And it is shame they did not use at least 512 MBV version... this is killing me with any mac computer, they always, ALWAYS use the lowest posible VRAM configs (except the high end imacs)...
 
Register on MacRumors! This sidebar will go away, and you'll see fewer ads.