The new iMac is a lot noisier!

[mathemabeat]

I am just curious why did you not consider the i5 3.5 version which I believe will be must cooler yet with similar CPU performance to the i5 3.8 according GeekBench result.

Mostly because I wanted to pick something up at retail and not BTO a machine. Plus the retail 3.5 machine has the kinda crappy 1tb fusion drive and 575 graphics. I wanted the 2tb fusion and 580 graphics.

And I don't believe you can select the Radeon Pro 580 with the 3.5ghz i5.

 

[cynics]

I am just curious why did you not consider the i5 3.5 version which I believe will be must cooler yet with similar CPU performance to the i5 3.8 according GeekBench result.

GeekBench doesn't always translate to real world use very well. Userbenchmark gives the 7600K an 8% advantage which is in direct correlation to its 8% higher base frequency. Basically certain people with certain task can benefit from that higher frequency.

An unlocked CPU in a Mac doesn't make a lot of sense but with that CPU option you get a faster GPU as well.

 

[Glideslope]

I run X-Plane 11 which actually can be very taxing on machines. With settings near or at max the fan would very quietly cycle a bit...but mostly it stayed quite....have not run the simulator for hours yet...but the performance is super and the Mac is behaving as it should. Took about 6 hours to transfer data from my old iMac but everything came across perfectly...not reinstalls needed...that is very cool.

Yea. I'm looking forward to Migration Assistant with the 2008 MP to the new iMac next week. 6 hrs was my target guess.

 

[user1234]

I run X-Plane 11 which actually can be very taxing on machines. With settings near or at max the fan would very quietly cycle a bit...but mostly it stayed quite....have not run the simulator for hours yet...but the performance is super and the Mac is behaving as it should. Took about 6 hours to transfer data from my old iMac but everything came across perfectly...not reinstalls needed...that is very cool.

Have you checked if X-Plane 11 is GPU or CPU limited on this machine? I'm thinking about getting the 2017 iMac but I'm having a hard time deciding between the i5 and the i7. I'm one of those users that can never have enough performance, but unfortunately my wallet doesn't currently reflect that too well

 

[trsblader]

Doing some transcoding here today on my 2011 21.5" (2.7 i5). The Cpu is between 156-159 and the cpu fan goes back and forth between 1900 and 2000 rpms. The HDD fan is at 1500, and the ODD is still idle. House is silent other than a little background noise from the tv downstairs and I cannot hear this thing over the external spinner drive sitting next to my computer. I have to put my ear right up to it before I can definitively say I hear it over the hard drive. These numbers are over the past hour or so. It will be nice not to have such high temps but I sure am going to miss the quietness of this machine when I upgrade to my 2017. I also will not miss one transcode taking almost 3 hours.

 

[propower]

That's maybe what I'd recommend in your situation or else a mid-tier i5, but you seemed set on the i7 with Turbo turned off. What changed your mind?

Update on update - after much thought - can't go with less than 1TB SSD. Not a big $$ difference then to stay with the i7/580. Going to try a week of recording work on it and then decide whether to keep or switch .

 

[macryan1]

Have you checked if X-Plane 11 is GPU or CPU limited on this machine? I'm thinking about getting the 2017 iMac but I'm having a hard time deciding between the i5 and the i7. I'm one of those users that can never have enough performance, but unfortunately my wallet doesn't currently reflect that too well

Just to follow up a bit....took a flight in X-Plane 11 from Portsmouth NH to Boston with settings nearly maxed out. About an hour of flying time in a small plane. Temps for cpu and gpu were in the mid 60s˚c. fan speeds in the mid 1200s. felt all around the back of the iMac and it really was just a touch warm...with my 2011 the top back of the mac was actually hot to the touch.

Couldn't be more pleased....with the settings I had I got in the high 20s fps in downtown Boston. That's good!

 

[EugW]

Confirmed propower's observation. If you install Turbo Boost Switcher for use with the 4.2 GHz Core i7 and Radeon Pro 580 to turn off Turbo Boost, it's harder to get the fans to ramp up. But if you max out the CPU (such as with software 4K HEVC decoding) you can still eventually get the fans to ramp up. It just takes longer.

For interest's sake, I ran Geekbench 4.1 with Turbo Boost turned off. Multicore speed was in the same ballpark, but single-core speed was a bit slower.

Turbo Boost On: 5821 / 20046
Turbo Boost Off: 5491 / 19394

5891/5821 = 1.060
20046/19394 = 1.033

Note that Turbo Boost on this chip is up to 4.5 GHz for single-core and up to 4.4 GHz for multi-core.

4.5/4.2 = 1.071
4.4/4.2 = 1.048

So, these numbers do kind of make sense. Obviously, I feel no difference whatsoever in the UI with Turbo Boost turned off.

So, if you're planning on doing some long hardcore h.265 encodes with Handbrake, don't expect the 4.2 GHz Core i7 to be silent even with Turbo Boost turned off. I think if you wanted it to be near silent, or else at least quiet for longer periods of time, you'd have to shut off HyperThreading.

IOW, if you want some decent performance, and you want a quiet machine for long periods of time while the CPU is maxed out, then maybe you should get the i5-7600K and install Turbo Boost Switcher on that. You'd be stuck at 3.8 GHz and there would be no HyperThreading. The problem there though of course is you wouldn't get the extra 30% performance you can see with the i7-7700K.

BTW, I was playing with a 3.8 GHz i5 with Radeon Pro 580 at the Apple Store. It's very nice too of course. However, the Fusion drive it had really slowed it down it seems for certain actions. For example, when scrolling quickly through pictures in their demo library in Photos very quickly, it would take a long time for certain pictures to display when you stopped to look at them. Up to several seconds. On the Core i7 4.2 I have an SSD, and it's either near instantaneous or else up to 0.5 seconds. I know it's not a CPU issue, because it's actually just as fast on a 15" MacBook Pro I tested in the store.

 

[mathemabeat]

Excellent findings!



Regarding you thoughts on the Fusion drive: I agree that for large amounts of data or anything the OS deems necessary to stick on the HD portion of the Fusion drive, it kinda sucks. I personally keep my OS drive on the light side. Just OS and applications. Usually well under 120 gigs. I am not sure if by doing that macOS keeps everything on the SSD, or if it still moves some stuff over to the hard drive. I keep my large working projects on a external Type C SSD and other large bulk storage on a gigabit connected NAS in the closet. 20TB FTW.


Speaking of Fusion drive, are there any utilities that can determine on which part of the Fusion drive files are stored? Would be interested to see where macOS stores certain data at.

 

[EugW]

Excellent findings!



Regarding you thoughts on the Fusion drive: I agree that for large amounts of data or anything the OS deems necessary to stick on the HD portion of the Fusion drive, it kinda sucks. I personally keep my OS drive on the light side. Just OS and applications. Usually well under 120 gigs. I am not sure if by doing that macOS keeps everything on the SSD, or if it still moves some stuff over to the hard drive. I keep my large working projects on a external Type C SSD and other large bulk storage on a gigabit connected NAS in the closet. 20TB FTW.

If that's the case then it makes sense to forego the Fusion drive and get a 256 GB SSD. Note also that the 1 TB Fusion drive only has a 24 GB SSD. You need to get the 2 TB Fusion drive to get the 128 GB SSD.

It should be noted that a 2 TB Fusion Drive upgrade (from 1 TB) costs more than the upgrade to the 256 GB SSD. However, IMO, for many people the 512 GB SSD may be the sweet spot, with external storage as needed.

 

[MMcCraryNJ]

Kinda off-topic, but one thing to keep in mind is that while you can always get external SSD storage, it will not even come close to the internal PCIe speeds that these drives are running. We're talking 400MB/s read on an external SSD vs 2400MB/s on the internal.

My advice is to buy pure SSD storage for the iMac, the highest you can comfortably afford and justify, and if you need more, use external drives at that point.

 

[BuCkDoG]

If that's the case then it makes sense to forego the Fusion drive and get a 256 GB SSD. Note also that the 1 TB Fusion drive only has a 24 GB SSD. You need to get the 2 TB Fusion drive to get the 128 GB SSD.

It should be noted that a 2 TB Fusion Drive upgrade (from 1 TB) costs more than the upgrade to the 256 GB SSD. However, IMO, for many people the 512 GB SSD may be the sweet spot, with external storage as needed.

That's what I got is the 512 SSD. I have the i7 580 and with the 512 SSD and it's perfect.

 

[mathemabeat]

If that's the case then it makes sense to forego the Fusion drive and get a 256 GB SSD. Note also that the 1 TB Fusion drive only has a 24 GB SSD. You need to get the 2 TB Fusion drive to get the 128 GB SSD.

It should be noted that a 2 TB Fusion Drive upgrade (from 1 TB) costs more than the upgrade to the 256 GB SSD. However, IMO, for many people the 512 GB SSD may be the sweet spot, with external storage as needed.

I have the 2TB Fusion drive in my machine, but excellent reminder.

I wish Apple would dump the Fusion drive in the 27 inch models. Or at least mid and upper end retail configs. I see keeping in the base 27 inch, but it makes no sense in the pricier machines.

I bought my machine retail, otherwise had I done a BTO I certainly would have opted for either a 256 or 512gig SSD.

 

[Chancha]

That's what I got is the 512 SSD. I have the i7 580 and with the 512 SSD and it's perfect.

The 512 BTO upgrade is hitting a pretty sweet spot in price-to-performance ratio, 256 in comparison is a rather limiting option with minimal savings.

Since USB3 ext storage is quite cheap nowadays, for iMac desktop usage the best setup is internal SSD for OS/apps/active project data, then ext HDD for pernemant data/archive/TimeMachine.

 

[BuCkDoG]

The 512 BTO upgrade is hitting a pretty sweet spot in price-to-performance ratio, 256 in comparison is a rather limiting option with minimal savings.

Since USB3 ext storage is quite cheap nowadays, for iMac desktop usage the best setup is internal SSD for OS/apps/active project data, then ext HDD for pernemant data/archive/TimeMachine.

I completely agree. The external storage is cheap enough to warrant the savings from the additional cost of the internal SSD upgrade. Sure it would be nice to have 1 TB but for my needs I really don't need that much SSD space and can get away with an external SSD for any additional space needed.

 

[fokmik]

i wonder if somebody had or has the 2014 i7+m295x and now is using the i7+580 how they compare, they still runs hot and loud like the m295x?

 

[enigma2k]

Is the 7600 3.8 Ghz i5 Apple is using a 65W or a 93W chip? From what I can see it is the "K" version also using 93W. So is it really more silent than the i7 that is using the same wattage?

 

[Chancha]

Is the 7600 3.8 Ghz i5 Apple is using a 65W or a 93W chip? From what I can see it is the "K" version also using 93W. So is it really more silent than the i7 that is using the same wattage?

"base config" is i5-7500 @ 3.4GHz
"mid config" is i5-7600 @ 3.5GHz
"top config" is i5-7600k @ 3.8GHz
and then BTO is i7-7700k @ 4.2GHz

 

[Falcon80]

Is the 7600 3.8 Ghz i5 Apple is using a 65W or a 93W chip? From what I can see it is the "K" version also using 93W. So is it really more silent than the i7 that is using the same wattage?

It is not as noisy due to the lesser heat (lower clock speed + no hyperthreading)

 

[Smoothie]

Is the 7600 3.8 Ghz i5 Apple is using a 65W or a 93W chip? From what I can see it is the "K" version also using 93W. So is it really more silent than the i7 that is using the same wattage?

The 7600K and the 7700K are both rated at 91 watts for TDP. TDP is not a measurement of how much power a chip uses. It's a rough measure of how much power is generated in heat by the chip so that a computer designer can come up with a good cooling solution. (There is some correlation between power used and heat dissipated.) The problem is that TDP is somewhat arbitrary, since chip manufacturers often assign the same TDP to a series of chips with different clock speeds and the existence or non-existence of hyperthreading. Clock speed and hyperthreading are major contributors to heat generation.

Although the 7600K and the 7700K have the same TDP according to Intel, the 7600K runs cooler than the 7700K. This is shown in tests where the chips are subjected to load testing. (The 7700K has a higher base clock speed and it has hyperthreading.) Less heat generation means lower fan speed. It's possible that Intel lumped the 7600K and the 7700K together regarding TDP because the "K" signifies that the chips have an unlocked multiplier which allows a user to overclock the chips. Apple does not provide for overclocking.

 

[Zapp21]

The 7600K and the 7700K are both rated at 91 watts for TDP. TDP is not a measurement of how much power a chip uses. It's a rough measure of how much power is generated in heat by the chip so that a computer designer can come up with a good cooling solution. (There is some correlation between power used and heat dissipated.) The problem is that TDP is somewhat arbitrary, since chip manufacturers often assign the same TDP to a series of chips with different clock speeds and the existence or non-existence of hyperthreading. Clock speed and hyperthreading are major contributors to heat generation.

Although the 7600K and the 7700K have the same TDP according to Intel, the 7600K runs cooler than the 7700K. This is shown in tests where the chips are subjected to load testing. (The 7700K has a higher base clock speed and it has hyperthreading.) Less heat generation means lower fan speed. It's possible that Intel lumped the 7600K and the 7700K together regarding TDP because the "K" signifies that the chips have an unlocked multiplier which allows a user to overclock the chips. Apple does not provide for overclocking.

Right! What I don't understand though is the significantly higher temperature/power consumption of the i5-7600K compared to the i5-7600.

The single digit performance increase of the 7600K means 30 more Watts in power consumption and 20 more degrees in temperature at full load. (according to Tom's Hardware)

 

[badlydrawnboy]

You all seem pretty knowledgable about this stuff so wondering if you can help. I have a late 2014 i7 4.0 Ghz with 512 GB SSD. The fan is audible a lot more often than I think it should be. I don't do any video editing or (at least what I think would be) processor-intensive stuff. I do have a podcast and produce a lot of YouTube type videos using Screenflow. I've found that even having Screenflow open, without recording anything, will kick on the fans. Sometimes the same thing happens with GarageBand.

I'm actually considering a new iMac with the i5 specifically to have less noise. Seems like the performance would be about similar to my current 2014 i7, which would be fine—I've never had a complaint about performance, with the possible exception of switching from Library to Develop module in Lightroom.

Or maybe I just need to do a clean install or something? Any reason you can think of why the fan would kick on more than it should, and how to troubleshoot that?

 

[DisMyMac]

Right! What I don't understand though is the significantly higher temperature/power consumption of the i5-7600K compared to the i5-7600.

The single digit performance increase of the 7600K means 30 more Watts in power consumption and 20 more degrees in temperature at full load. (according to Tom's Hardware)

I thought that too. I just want to save electricity when it is idle though, not worried about full load.

 

[EugW]

You all seem pretty knowledgable about this stuff so wondering if you can help. I have a late 2014 i7 4.0 Ghz with 512 GB SSD. The fan is audible a lot more often than I think it should be. I don't do any video editing or (at least what I think would be) processor-intensive stuff. I do have a podcast and produce a lot of YouTube type videos using Screenflow. I've found that even having Screenflow open, without recording anything, will kick on the fans. Sometimes the same thing happens with GarageBand.

I'm actually considering a new iMac with the i5 specifically to have less noise. Seems like the performance would be about similar to my current 2014 i7, which would be fine—I've never had a complaint about performance, with the possible exception of switching from Library to Develop module in Lightroom.

Or maybe I just need to do a clean install or something? Any reason you can think of why the fan would kick on more than it should, and how to troubleshoot that?

According to Apple, the new Core i7 iMacs are significantly lower power than the older Core i7 iMacs.

https://support.apple.com/en-ca/HT201918

iMac Late 2014 27" Core i7 4 GHz Radeon M295X
Idle: 70W
CPU Max: 288W

iMac Late 2015 27" Core i7 4 GHz Radeon M390
Idle: 63W
CPU Max: 240W

iMac Mid 2017 27" Core i7 4.2 GHz Radeon 580
Idle: 71W
CPU Max: 217W

1. Power consumption data (watts) is measured from the wall power source and includes all power supply and system losses. Additional correction isn't needed.

2. CPU Max is defined as running a compute-intensive test application that maximizes processor usage and therefore power consumption.

3. These numbers reflect a 23°C (73.4°F) ambient running environment. Increased ambient temperatures requires faster fan speeds which increases power consumption. At 35°C (95°F), 50W should be added to reflect increased power consumption.

---

These numbers are somewhat difficult to compare as I suspect some of the increased power usage in the older models is likely related to the GPU differences, but nonetheless it would seem the Core i7 iMac in 2017 may be cooler and overall (including GPU changes), the 2017 iMacs are likely quieter than the Core i7 iMac in 2014. According to Apple, the overall system idle power for your 2014 and the current i7 is the same, but at CPU Max, your 2014 uses 70 Watts more power. However, if you want to be absolutely sure a new iMac is very quiet at all times, you might want to consider the i5-7600 non-K.

BTW, take a look at this one:

iMac Mid 2010 27" 3.6 GHz Core i5 Radeon 5670
Idle: 145W
CPU Max: 365W

Ouch, that's a lot of power, even at idle.

That's a dual-core though. They don't list my 2010 quad core i7.

 

[koyoot]

KabyLake CPUs are actually using slightly more power than Skylake CPUs, according to reviews. So this cannot be only CPU load power consumption.

217W? CPU alone consumes around 90W. 127W for the rest of the system? How much less power must consume the display, then?

OTOH:
iMac (21.5-inch, Late 2013)
21.5-inch display, 2.7GHz Intel Core i5, 8GB 1600MHz DDR3 SDRAM, 1TB Serial ATA Hard Drive, Intel Iris Pro Graphics
Power Consumption Thermal Output
Idle 38W
CPU Max 91W

This is 65W TDP CPU. It is actually possible, that Radeon Pro 580 is 100W TDP GPU.