MP 1,1-5,1 Is the RX 580 bottlenecked by the Mac Pro 3, 1

[TechInit]

Yes, I am well aware it's not possible to run macOS with the RX XXX due to the lack of the SSE4.2 thing, however, I use Windows a fair bit, and there are methods to fake the SSE4.2 instruction set by using MouSSE. All I want to know is if its worth getting an RX 580, and using it in my Mac Pro 3 1

Thanks!

 

[tsialex]

Yes, I am well aware it's not possible to run macOS with the RX XXX due to the lack of the SSE4.2 thing, however, I use Windows a fair bit, and there are methods to fake the SSE4.2 instruction set by using MouSSE. All I want to know is if its worth getting an RX 580, and using it in my Mac Pro 3 1

Thanks!

Today anything is bottlenecked by a C2D era Xeon, but make your upgrade decisions not based on your current MP3,1, buy everything that you can later re-use on a early-2009 or mid-2010/mid-2012 Mac Pro build.

It's worth to get a RX 580, even bottlenecked now with your 2008 Mac Pro.

 

[Ludacrisvp]

No regrets with my 3,1 and the rx 580. World of difference vs the previous cards I’ve used in it. (GTX650/HD5770/HD5450/8800GT)

 

[BilduKat]

How well does the RX 580 work in a 3,1 ?
Can it run Mojave?
I have a Nvidia 750ti in mine currently, will a RX580 perform better?
If I was to install a RX580 what would I need?
I have a dual 6-pin to 8-pin cable. Is that enough for power?

Thanks in advance!!

Jason

 

[fendersrule]

Yes, it is bottlenecked.

 

[Ludacrisvp]

Yes, it is bottlenecked in every cMP (both 3,1 and 5,1), but its so minor it really doesn't matter.

I have had no complaints using the RX580 in mine.

The massive win is the ability to playback and encode h264/h265/HEVC video streams using the GPU using the AMD hardware acceleration feature that doesn't exist for Nvidia in macOS. This allows the 3,1 to playback the highest bitrate encoded 4k video from the jellyfish site with 0 dropped frames, without this the 3,1 just shows a still image.

Jellyfish Bitrate Test Files

A collection of .mkv video clips encoded at various bitrates; useful for testing the network streaming and playback performance of media streamers & HTPCs.

jell.yfish.us

You won't be playing this video back in macOS in that 3,1 using any nvidia card, but this RX580 can do it flawlessly and not even bother the CPU in the 3,1 to do it.

jellyfish-400-mbps-4k-uhd-hevc-10bit.mkv

400 Mbps

3840x2160

HEVC

Main10

6.1

High

1.4 GB

The bottleneck isn't very big, but also probably depends on what you use it for.

Compared a Dell T5810 hackintosh (6core 12thread @3.5Ghz xeon) with the RX580 8Gb Red Devil which is using PCIe 3.0 (the iMacPro1,1 in the results) to the cMP 3,1 (the "vmware7,1" OpenCore booted cMP 3,1) and the scores aren't much different considering the cMP 3,1 is also using a RX 580 4Gb Pulse.

Basically 5k point loss on metal benchmark

VMware7,1 - Geekbench

Benchmark results for a VMware7,1 with an Intel Xeon X5482 processor.

browser.geekbench.com

iMacPro1,1 - Geekbench

Benchmark results for an iMacPro1,1 with an Intel Xeon E5-1650 v3 processor.

browser.geekbench.com

And about 4k loss on OpenCL benchmark

iMacPro1,1 - Geekbench

Benchmark results for an iMacPro1,1 with an Intel Xeon E5-1650 v3 processor.

browser.geekbench.com

VMware7,1 - Geekbench

Benchmark results for a VMware7,1 with an Intel Xeon X5482 processor.

browser.geekbench.com

 

[fendersrule]

Yes, it is bottlenecked in every cMP (both 3,1 and 5,1), but its so minor it really doesn't matter.

Well, not so fast there. I know maybe gaming wasn't the original question, but the 3,1 and 5,1 are not even in the same league. The 5,1 @ 3.46Ghz is not bottlenecked with a RX580 for gaming at 1080p and onwards. The 3,1 is DEFINITELY bottlenecked with a 1060 6B (similar to an RX580) at 1080p (confirmed).

Truth is that the first gen i7 (5,1) still holds up very well even today. The 3,1 is a core2duo era machine, and they are very slow by today's standards. Upgrading from a 3,1 2.8GHz quad core to a 5,1 2.8GHz quad core was a night and day difference. I would not put the 3,1 and the 5,1 on the same playing field for general OS use and gaming as I have experience with both.

Upgrading from a RX580 to a GTX 1070 netted me ~30% improvements across the board in all benches that I ran. This is evidence that the Mac Pro 5,1 does not bottleneck the GTX 1070 since the percentage improvements are matching what is known. My best estimations is that a modern CPU is going to push the GTX 1070 roughly around 8-15%ish faster at 1080p resolutions, give or take, depending on that game. That's not enough to say it's "bottlenecked".

Bottlenecked is taking a GTX 1060 out of a 3,1 that couldn't even push 50 FPS in TF2 1080p and sticking it in a modern PC and all of a sudden it gets 280 FPS. Confirmed.

Anyways, carry on with the encoding discussion because there may be some truth there. Just didn't want to let that quote get taken out of context. I do not miss my 3,1. I am glad to be rid of all Core2Duo era machines at this point.