iMac 2020 Reviews And Benchmarks - The Last Intel iMac

[GCtravellerguy]

It's monoplisation of the worst kind and it's un-American.

Yeah I concur with the term UnAmerican but I feel like the Aussie market mirrors you guys a lot and its what we would call a "**** MOVE" I believe in freedom to do what you want but profits at the detriment to your core customers is pushing an agenda I would liken to a Dictatorship

 

[nihil0]

I am well aware of that, it's poorly optimised even for Windows. Yet it should run much better with throwing better compute, faster r/w rates and a beefier CPU at it.

It will run better for sure. But it will still be underperforming.

 

[John90976]

Yeah I concur with the term UnAmerican but I feel like the Aussie market mirrors you guys a lot and its what we would call a "**** MOVE" I believe in freedom to do what you want but profits at the detriment to your core customers is pushing an agenda I would liken to a Dictatorship

Not to make it any more political than it is, but if your government doesn't certify your right to repair and make sure companies operating within its grasp follow that guidance, then the power is in the hands of the corporation to do as they please, sometimes it works out best for the consumer, in the case of Apple, we needed right to repair laws revitalized here in the US over a decade ago. That would've been helpful. Apple should be penalized alone for the deterrent methods they use on top of a general lack of repair-ability, Apple is user hostile if your brain goes too far out of their box, same with the app store. Does not mean I don't like Apple's fun box of ideas, I love it, it's just artificially limiting and user hostile.

 

[JMacHack]

Not to make it any more political than it is, but if your government doesn't certify your right to repair and make sure companies operating within its grasp follow that guidance, then the power is in the hands of the corporation to do as they please, sometimes it works out best for the consumer, in the case of Apple, we needed right to repair laws revitalized here in the US over a decade ago. That would've been helpful. Apple should be penalized alone for the deterrent methods they use on top of a general lack of repair-ability, Apple is user hostile if you brain goes too far out of their box, same with the app store. Does not mean I don't like Apple's fun box of ideas, I love it, it's just artificially limiting and user hostile.

We have "right to repair" laws already, it's just that they're narrowly defined to automobiles. Frankly it should be expanded to all devices.

Also, I should note, that there's nothing in "Right to Repair" law that says companies have to make their products easy to repair. It certainly wouldn't fix the T2 chip issue, as Apple could just solder the memory chips directly to the same PCB as the T2 chip itself. (which, to be fair, I think would be a good solution, and have that particular PCB be removable, almost as if it was a drive itself.)

Actually I think its very American. Many US family fortunes were built on the profits from monopolization of markets.

I don't plan on continuing this argument further over what is or is not "American" (This isn't PRSI and I don't want this to turn into the ****-flinging fest that it is), but depending on how you define it ("American" in the sense that it's consistent with our history, or "American" in the sense that it's consistent to our ideals) it could mean a number of different things.

 

[John90976]

Got some decent benchmarks out of this solid review of maxed out model. 5700XT generally matches 2070 Super so its a bit of a shame it's that it looks likes its anywhere from 15-20% weaker due to lower power.

 

[Homy]

iMac 10-core 5700 XT faster in almost everything than iMac Pro Vega 64!

2020 27-inch iMac review: A classic Mac for the end of an era

With Apple Silicon around the corner, is this Intel-based Mac refresh a good buy?

arstechnica.com

 

[Homy]

5700 XT eGPU half as good as 5300 dGPU in Unigine!

 

[quagmire]

So Max did more benchmarking with the 5500 XT. Some interesting results. In the two GFXBench results he showed and the Unigine test with the framerate only, the 5500 XT is not far behind my Vega 48. In the Unigine test, the 5500 XT pulls 64 FPS while my Vega 48 does 67 FPS. In GFXBench, the 1440p Aztec Ruins test results are pretty much equal at 136 FPS( 5500) and 137 FPS( Vega 48). The Manhattan 1440p test has the 5500 at 267 FPS and my Vega 48 pulls 287 FPS so only 9% off pace there for the 5500 XT.

I am hardly an expert on benchmarks, but keeping track of all these benchmarks results so far seem to show some inconsistencies. Geekbench Metal scores doesn't have the 5500 XT close at all to the Vega 48. The Unigine FPS wise show it close and the Aztec Ruins tests has it close. The Manhattan test has it 9% slower than the Vega 48.

 

[John90976]

The 5700XT with 16GB in this iMac is killer, it has surpassed my expectations and actually runs Red Dead Redemption 2 in 4K on high with no issues, pure beauty, new details never before seen because the game was designed for 4K. I don't have a frame monitor on but it's butter and my house mate with a 2070 Super is smitten at this things performance as he can't run RDR2 in 4K, his GPU uses 3x the power and is very loud. I'm very pleased with this, and for me the i7 is wonderful overkill.

 

[sco]

Got some decent benchmarks out of this solid review of maxed out model. 5700XT generally matches 2070 Super so its a bit of a shame it's that it looks likes its anywhere from 15-20% weaker due to lower power.

The comparisons between the iMac and MBPro 16 in this one were interesting. Also the discussion about base vs higher spec fan noise, GPU power usage behaviour (causing CPU power to be limited).

 

[JMacHack]

I am hardly an expert on benchmarks, but keeping track of all these benchmarks results so far seem to show some inconsistencies. Geekbench Metal scores doesn't have the 5500 XT close at all to the Vega 48. The Unigine FPS wise show it close and the Aztec Ruins tests has it close. The Manhattan test has it 9% slower than the Vega 48.

I won't pretend to be an expert, but from what little I understand is that Metal is better about leveraging Vega's compute abilities than other graphics benchmarks. Somehow Vega was a number-crunching monster while being weak in traditional graphics tasks.

 

[triangletechie]

I've had my 2020 27" iMac for 2 days and it just restarted due to a kernel panic error. Off to a great start.

 

[Homy]

Comparison between lots of Macs in French

 

[Darajavahus]

Is it not possible to delete post? Coz I can't find it

 

[oneafour]

just finished upgrading my ram to 64gb, removed the previous 8gb to ensure dual channel is working,

i9 10core, 64gb ram, 5500xt
and my Geekbench score is:
Single Core: 1160
Multi Core: 9241

Does this look right?

I removed the two 4gb sticks from apple and replaced them with the 32gb sticks. From the Memory tab in About This Mac, it is the bottom two memory slots and they've been detected as 32 + 32 @ 2667Mhz.


edit:
reran
Single core: 1284
Multi: 9784

 

[SWolfe]

a little low but within range. you can search the i9-10910 result on Geekbench Browser. Some things that seem to help with a higher score. Make sure you have no other programs running, Allow the Mac to sit for a while and let spotlight index. Make sure to check activity monitor for any background tasks. Some people also run up there fans to pre-cool and start off with a CPU temp of around 40C and also unplug any secondary monitors (but these last 2 feels like cheating to me)

 

[KrazyKanuck]

just finished upgrading my ram to 64gb, removed the previous 8gb to ensure dual channel is working,

i9 10core, 64gb ram, 5500xt
and my Geekbench score is:
Single Core: 1160
Multi Core: 9241

Does this look right?

That seems a little off. Your single core should be closer to 1300, and multi-core should be nearer 10,000. What is your RAM configuration? Are there any other system processes currently active (ie. is Spotlight indexing)?

 

[SWolfe]

the highest I have seen so far was approx 1390s Single and 10400s but they had 128GB RAM and don't know their exact config was except 2020 iMac i9

 

[oneafour]

ill try rerunning the tests later tonight but I'm gonna download all my apps and stuff now. I didn't check if spotlight was indexing, but I wouldn't be surprised as I ran it before I installed/done anything else.

 

[KrazyKanuck]

ill try rerunning the tests later tonight but I'm gonna download all my apps and stuff now. I didn't check if spotlight was indexing, but I wouldn't be surprised as I ran it before I installed/done anything else.

Sounds like a plan! Look at Activity Monitor before you run the test. Your CPU usage should be in the range of 1-2% for background activities, then you can run Geekbench.

 

[fireguy286]

PERFORMANCE ON THE 27-INCH IMAC
It’s fast! It ought to be, as the model I am testing is the Core i9 model with 32GB of RAM and a Radeon Pro 5700 XT graphics card. With the nano texture screen option and a 1TB SSD, the machine Apple provided for review would retail for around $4,500. It’s 30 percent faster at one of our standard 4K Adobe Premiere Pro exports than my 2017 Core i7 iMac with a Radeon Pro 580.

But really, the performance on an iMac shouldn’t be an X-factor. How these machines operate — even under heavy thermal loads — is pretty much a known quantity. The good news is that, so far, the new 10th Gen Intel chips fare just fine inside this very familiar chassis.

Apple iMac 27-inch (2020) review: new webcam, new screen option, same iMac

The nano texture display looks really good, but is it worth $500?

www.theverge.com

 

[KrazyKanuck]

PERFORMANCE ON THE 27-INCH IMAC
It’s fast! It ought to be, as the model I am testing is the Core i9 model with 32GB of RAM and a Radeon Pro 5700 XT graphics card. With the nano texture screen option and a 1TB SSD, the machine Apple provided for review would retail for around $4,500. It’s 30 percent faster at one of our standard 4K Adobe Premiere Pro exports than my 2017 Core i7 iMac with a Radeon Pro 580.

But really, the performance on an iMac shouldn’t be an X-factor. How these machines operate — even under heavy thermal loads — is pretty much a known quantity. The good news is that, so far, the new 10th Gen Intel chips fare just fine inside this very familiar chassis.

Apple iMac 27-inch (2020) review: new webcam, new screen option, same iMac

The nano texture display looks really good, but is it worth $500?

www.theverge.com

Thank goodness. Though I would have enjoyed seeing them open it up to reapply fresh thermal compound. Lots of thermal compound... 🤪

 

[oneafour]

just finished upgrading my ram to 64gb, removed the previous 8gb to ensure dual channel is working,

i9 10core, 64gb ram, 5500xt
and my Geekbench score is:
Single Core: 1160
Multi Core: 9241

Does this look right?

I removed the two 4gb sticks from apple and replaced them with the 32gb sticks. From the Memory tab in About This Mac, it is the bottom two memory slots and they've been detected as 32 + 32 @ 2667Mhz.


edit:
reran
Single core: 1284
Multi: 9784

Also ran metal with a 5500xt:
43013

(I don't recall benchmarks with this cpu+gpu+ram combo hopefully it helps someone decide)

 

[SWolfe]

That’s my exact config right down to the 64g ram so thanks.

 

[Homy]

PERFORMANCE ON THE 27-INCH IMAC
It’s fast! It ought to be, as the model I am testing is the Core i9 model with 32GB of RAM and a Radeon Pro 5700 XT graphics card.

Apple iMac 27-inch (2020) review: new webcam, new screen option, same iMac

The nano texture display looks really good, but is it worth $500?

www.theverge.com

I wonder how he knows it's fast. His only reason is "it ought to be" because of i9 and 5700 XT. Makes me wonder if he even has used the computer in some apps...