How's the love for G5s with a single 1.8 GHz cpu?

[bobesch]

Everything appears to be in line. Here is the benchmark with a single G4 933GHz Quick Silver, 512 MB RAM, OS X 10.4.?:

File: test.mp4
Sunday, 06:51:04: Export Stream started
Sunday, 06:51:39: Encoding
Sunday, 06:52:37: Encoding
Sunday, 06:54:51: Encoding
Sunday, 06:57:21: Encoding
Sunday, 06:59:52: Encoding
Sunday, 07:02:19: Encoding
Sunday, 07:04:42: Encoding
Sunday, 07:07:04: Encoding
Sunday, 07:09:34: Encoding
Sunday, 07:12:04: Encoding
Sunday, 07:14:34: Encoding
Data rate 4.68 Mbps (4.55 video, 0.13 audio)

Time to complete: 00:22:49

And my 2.7GHz Core i7 MacBook Pro, 16GB RAM, 768GB SSD, OS X 10.9.5:

File: test.mp4
Sunday, 07:37:32: Export Stream started
Sunday, 07:37:34: Encoding
Sunday, 07:37:37: Encoding
Sunday, 07:37:46: Encoding
Data rate 4.79 Mbps (4.67 video, 0.13 audio)
Sunday, 07:37:48: Movie completed

Time to complete: 00:00:16

If you compare the time spent to do that job on both the machines and if you take into account the power-consumption of both the devices I wouldn't recommend a G5 for daily use but to go for e.g. a core2duo >2GHz MacBook. The higher price of such a device certainly will be compensated by your energy-savings... (on the other hand you will have less costs for heating...)
It's a pity - despite of their gorgeous design those PowerMacs are dinosaurs concerning power-consumption.
Look a the new iBook - it would be funny to upgrade the chassis of the G5 to that tiny logic board.
So much space left and the space 'n saved energy would be sufficient for a nice house bar and making ice cubes
The CubeG4 and the PPC-"Books" are of another kind and lower-performance/longer running time doesn't have such a big impact on your energy bill.
Sorry about that nice machines...

 

[996085]

If you compare the time spent to do that job on both the machines and if you take into account the power-consumption of both the devices I wouldn't recommend a G5 for daily use but to go for e.g. a core2duo >2GHz MacBook. The higher price of such a device certainly will be compensated by your energy-savings... (on the other hand you will have less costs for heating...)
It's a pity - despite of their gorgeous design those PowerMacs are dinosaurs concerning power-consumption.
Look a the new iBook - it would be funny to upgrade the chassis of the G5 to that tiny logic board.
So much space left and the space 'n saved energy would be sufficient for a nice house bar and making ice cubes
The CubeG4 and the PPC-"Books" are of another kind and lower-performance/longer running time doesn't have such a big impact on your energy bill.
Sorry about that nice machines...

Funny that you would mention a 2.0GHz Core 2 Duo MacBook:

File: test.mp4

Sunday, 16:12:05: Export Stream started
Sunday, 16:12:17: Encoding
Sunday, 16:12:34: Encoding
Sunday, 16:13:11: Encoding
Sunday, 16:13:54: Encoding
Sunday, 16:14:38: Encoding
Sunday, 16:15:21: Encoding
Sunday, 16:16:05: Encoding
Sunday, 16:16:48: Encoding
Sunday, 16:17:31: Encoding
Sunday, 16:18:15: Encoding
Sunday, 16:18:35: Encoding
Data rate 6.85 Mbps (6.73 video, 0.13 audio)
Sunday, 16:18:38: Movie completed

Total time to complete: 00:06:33

Core 2 Duo @ 2.0GHz, 2GB RAM, OS X 10.6.8

 

[bobesch]

Funny that you would mention a 2.0GHz Core 2 Duo MacBook:

File: test.mp4
[...]
Movie completed

Total time to complete: 00:06:33

Core 2 Duo @ 2.0GHz, 2GB RAM, OS X 10.6.8

The Core2Duo machines iMac white (150-250€/$)/MacBook2008(250-300€/$) with 2.0-2.4GHz are the machine I could compare to the PM-G5 by my own experience.
It's really interesting to see your benchmarks about video-encryption:
i7 (2,7) : 00:16
c2d (2,0): 06:33
PM G4: 23:40
PM G5 1x1.8: 1/3-1/4 of the Quicksilver? (06:00)
So the PMG5 1x1,8 seems to be the same speed or most likely faster than "my choice" of c2duo. DC/QC might do the job within 03:00/01:30 (if this rough gessing isn't completely wrong...)
Chapeau! - Especially if compared with the i7.
Would be interesting to compare speed/time/power-consumption.
I presume my previous calculation about savier c2duo might be obsolete too ...
Make me curious to run a comparison
with my PM-G5 2x2.0 / MB2008alu c2d 2,4 / MBP i7 2,9
(I hope I'll find some spare time soon and like to include messuring power-consumption)

 

[996085]

The Core2Duo machines iMac white (150-250€/$)/MacBook2008(250-300€/$) with 2.0-2.4GHz are the machine I could compare to the PM-G5 by my own experience.
It's really interesting to see your benchmarks about video-encryption:
i7 (2,7) : 00:16
c2d (2,0): 06:33
PM G4: 23:40
PM G5 1x1.8: 1/3-1/4 of the Quicksilver? (06:00)
So the PMG5 1x1,8 seems to be the same speed or most likely faster than "my choice" of c2duo. DC/QC might do the job within 03:00/01:30 (if this rough gessing isn't completely wrong...)
Chapeau! - Especially if compared with the i7.
Would be interesting to compare speed/time/power-consumption.
I presume my previous calculation about savier c2duo might be obsolete too ...
Make me curious to run a comparison
with my PM-G5 2x2.0 / MB2008alu c2d 2,4 / MBP i7 2,9
(I hope I'll find some spare time soon and like to include messuring power-consumption)

I can help with some power consumption measurements too. I plug both the G5 and the MacBook into my Kill-A-Watt meter and performed two tests on each: One at idle and the other with the processor pegged at 100% using the Distributed Net client. Each reading was taken over an hours time (i.e. I ran the test for an hour each test). Here are the power consumption readings for each:

G5:

Idle: .1/Kwh
Working: .14/Kwh

MacBook:

Idle: .01/Kwh
Working: .03/Kwh

 

[bobesch]

I can help with some power consumption measurements too. I plug both the G5 and the MacBook into my Kill-A-Watt meter and performed two tests on each: One at idle and the other with the processor pegged at 100% using the Distributed Net client. Each reading was taken over an hours time (i.e. I ran the test for an hour each test). Here are the power consumption readings for each:

G5:
Idle: .1/Kwh
Working: .14/Kwh
MacBook:
Idle: .01/Kwh
Working: .03/Kwh

Huuh, very interesting data. Thanks for sharing them with us!
If my MacBook is standby 20/24h and I'm working on it 4/24h the difference in power-consumption is obviously enormous.
Here's my calculation:
with an average price of 0,25c/kW and the above usage of my standby-computer I have an extra expense of 200€ if I'd go with the G5 instead of the MacBook c2duo.
Or:
for each average hour of daily usage the G5 takes an extra of 10€/year;
for each average hour of daily standby there's an extra of 8€/year.
If that is for the single-core G5 it would be interesting how much more energy the Dual/Quad-Core G5 is taking.
At least a two years period of usage seems to save enough money to buy a decent 2nd-hand hardware update with less power consumption.
Or did I make any mistakes in my calculation...

 

[996085]

Huuh, very interesting data. Thanks for sharing them with us!
If my MacBook is standby 20/24h and I'm working on it 4/24h the difference in power-consumption is obviously enormous.
Here's my calculation:
with an average price of 0,25c/kW and the above usage of my standby-computer I have an extra expense of 200€ if I'd go with the G5 instead of the MacBook c2duo.
Or:
for each average hour of daily usage the G5 takes an extra of 10€/year;
for each average hour of daily standby there's an extra of 8€/year.
If that is for the single-core G5 it would be interesting how much more energy the Dual/Quad-Core G5 is taking.
At least a two years period of usage seems to save enough money to buy a decent 2nd-hand hardware update with less power consumption.
Or did I make any mistakes in my calculation...

I didn't check the math but it's obvious the MacBook is the wiser choice not only when it comes to speed but also power savings. None of this should come as a surprise as the MacBook is later technology compared to the G5. Especially when the G5 processor was known to be power hungry.

What did surprise me was how little additional power each consumed when it was working compared with idle. I was surprised the working number was slightly higher than idle...for both. Maybe distributed net isn't taxing enough (though I had Activity Monitor running and it showed the CPU cores pegged at 100%).

In the end the G5, being older technology, takes longer and consumes more power. I don't own my G5 because of its speed / light energy usage. I own it because it has a little nostalgia for me and I like the different processor architecture that comes with it.

 

[QSDP-User]

I can help with some power consumption measurements too. I plug both the G5 and the MacBook into my Kill-A-Watt meter and performed two tests on each: One at idle and the other with the processor pegged at 100% using the Distributed Net client. Each reading was taken over an hours time (i.e. I ran the test for an hour each test). Here are the power consumption readings for each:

G5:

Idle: .1/Kwh
Working: .14/Kwh

MacBook:

Idle: .01/Kwh
Working: .03/Kwh

pl595, do you still own the MacPro 2,1 you've mentioned in the MacPro forum?

Just for the fun of it, how about a similar Kill-A-Watt test?

Or if you own a MacPro 1,1...that instead.

Thanks.

 

[996085]

pl595, do you still own the MacPro 2,1 you've mentioned in the MacPro forum?

Just for the fun of it, how about a similar Kill-A-Watt test?

Or if you own a MacPro 1,1...that instead.

Thanks.

No, I no longer own either but I do own a 2010 Mac Pro. I can perform some power tests tomorrow if you'd like.

 

[QSDP-User]

No, I no longer own either but I do own a 2010 Mac Pro. I can perform some power tests tomorrow if you'd like.

Hmmm, ya know, now that I think about it,
later Mac Pros might take the power consumption comparison a bit out of context
for a PPC discussion that started with a single 1.8 G5
...though I appreciate the offer & I know how these forum threads always go off on tangents of their own

 

[bobesch]

Hmmm, ya know, now that I think about it,
later Mac Pros might take the power consumption comparison a bit out of context
for a PPC discussion that started with a single 1.8 G5
...though I appreciate the offer & I know how these forum threads always go off on tangents of their own

So back to the roots.
"Hows the love to go with a G5..."
Yeah, it's simply only love and nostalgia - not reason ...

 

[QSDP-User]

So back to the roots.
"Hows the love to go with a G5..."
Yeah, it's simply only love and nostalgia - not reason ...

Dammit Spock, I'm just a simple country tower user.
And G5s are not dead, Jim.

Calm down Bones.
And Spock, I'm not a simpering, devil-eared freak, whose father was a computer and his mother an encyclopedia.


Edit: I should add that this reply is NOT aimed at anyone in particular.
Hopefully its quotes are recognized from a nostalgic TV series & sarcastically
& somewhat humorously brings up a different POV.

Apologizes if taken the wrong way.

 

[bobesch]

Dammit Spock, I'm just a simple country tower user.
And G5s are not dead, Jim.

Calm down Bones.
And Spock, I'm not a simpering, devil-eared freak, whose father was a computer and his mother an encyclopedia.


Edit: I should add that this reply is NOT aimed at anyone in particular.
Hopefully its quotes are recognized from a nostalgic TV series & sarcastically
& somewhat humorously brings up a different POV.

Apologizes if taken the wrong way.

"Scotty, watch that uge G5 prototype right there in the background. It's frontplate was blown off by its overheated fans ... How much time do you need to fix it?"
"A day or two, Jim."
"Ok, so you've got ten minutes!"
Uhuuuu huhuhuhuhuhuhuhu huuuuuuu



"Lt.Uhura, powerUp-PPC to sol-5, lets get away from this Classic place"

 

[QSDP-User]

"Scotty, watch that uge G5 prototype right there in the background. It's frontplate was blown off by its overheated fans ... How much time do you need to fix it?"
"A day or two, Jim."
"Ok, so you've got ten minutes!"
Uhuuuu huhuhuhuhuhuhuhu huuuuuuu

"Lt.Uhura, powerUp-PPC to sol-5, lets get away from this Classic place"

Well, I was waiting for you to incorporate this:

Spock: Captain, you're asking me to work with equipment which is hardly very far ahead of stone knives and bearskins.

 

[996085]

Hmmm, ya know, now that I think about it,
later Mac Pros might take the power consumption comparison a bit out of context
for a PPC discussion that started with a single 1.8 G5
...though I appreciate the offer & I know how these forum threads always go off on tangents of their own

OK...I'm happy to run some numbers if you'd like.

 

[QSDP-User]

Well fellow PPC-ers,

I'm gonna post something that me friends pl595 & perhaps bobesch could agree with:
I bought that G5 2.0 GHz Dual for $25 - It was still available!

I'm posting from it.
PowerMac 7,3
5.5 GB RAM, 2x 300 GB HDD, Radeon 9650.
Very clean & quiet.

The seller, a most likable cat,
is a professional sound man & composer,
a dedicated Mac-user doing what he does best since the 1980s.

He seemed like one of those modest artists who doesn't really vocally articulate what & how he does,
but he does it extremely well.

He mentioned an era where he had to deal with external drives & expansion cases,
then next Macs that could contain more drives and expansion cards.

I asked him what he thought of the "Trash Can" Pro, which he uses.
Shrugging, he said it's super fast & works well in a competitive world with
clients who want the job done (yesterday, probably).
But also it also returns to the external expansion world (Thunderbolt).

I mentioned that I thought a primo professional could recoup the hardware expenses with one or two primo jobs.
He said well, hardware is just one of the expenses; you have to consider the studio rent, software, accounting, taxes, etc.
*snap* This is not mother's basement, duuuhh, LOL

Anyway,
I got a G5 Mama Burger (analogy to A & W Drive-in burger/fries size).
So I won't be getting the Baby Burger & I don't see myself getting the Papa Burger.

And thanks to all input into this thread.

 

[996085]

Well fellow PPC-ers,

I'm gonna post something that me friends pl595 & perhaps bobesch could agree with:
I bought that G5 2.0 GHz Dual for $25 - It was still available!

I'm posting from it.
PowerMac 7,3
5.5 GB RAM, 2x 300 GB HDD, Radeon 9650.
Very clean & quiet.

The seller, a most likable cat,
is a professional sound man & composer,
a dedicated Mac-user doing what he does best since the 1980s.

He seemed like one of those modest artists who doesn't really vocally articulate what & how he does,
but he does it extremely well.

He mentioned an era where he had to deal with external drives & expansion cases,
then next Macs that could contain more drives and expansion cards.

I asked him what he thought of the "Trash Can" Pro, which he uses.
Shrugging, he said it's super fast & works well in a competitive world with
clients who want the job done (yesterday, probably).
But also it also returns to the external expansion world (Thunderbolt).

I mentioned that I thought a primo professional could recoup the hardware expenses with one or two primo jobs.
He said well, hardware is just one of the expenses; you have to consider the studio rent, software, accounting, taxes, etc.
*snap* This is not mother's basement, duuuhh, LOL

Anyway,
I got a G5 Mama Burger (analogy to A & W Drive-in burger/fries size).
So I won't be getting the Baby Burger & I don't see myself getting the Papa Burger.

And thanks to all input into this thread.

Excellent find...at $25 it's difficult to pass up...even if it would have been a single core system. The single core systems are OK but the dual core felt so much smoother to me. Given the price difference, if any, between the two the dual core is the wise choice. I paid $100 for my 1.8GHz system. I could have purchased the dual 2.0GHz for $200. Given this was four years ago the system was long obsolete already and I knew paying the extra $100 for the dual core wasn't going to buy me anything (as it was more a play system). Of course I managed to find a dual 2.0GHz about two years ago for $65. I picked it up because it had an SSD in it and the SSD alone was worth the money.

Would you mind running the benchmark on your new system?

 

[QSDP-User]

Would you mind running the benchmark on your new system?

Sure, I ran it yesterday:

File: G5_11-24-15_test.mp4
Tuesday, 14:16:30: Export Stream started
Tuesday, 14:16:36: Encoding
Tuesday, 14:16:45: Encoding
Tuesday, 14:17:00: Encoding
Tuesday, 14:17:18: Encoding
Tuesday, 14:17:36: Encoding
Tuesday, 14:17:54: Encoding
Tuesday, 14:18:11: Encoding
Tuesday, 14:18:29: Encoding
Tuesday, 14:18:47: Encoding
Tuesday, 14:19:04: Encoding
Tuesday, 14:19:22: Encoding
Data rate 4.79 Mbps (4.66 video, 0.13 audio)
Tuesday, 14:19:25: Movie completed

14:19:25
14:16:30
--------
00:02:55

 

[996085]

Sure, I ran it yesterday:

File: G5_11-24-15_test.mp4
Tuesday, 14:16:30: Export Stream started
Tuesday, 14:16:36: Encoding
Tuesday, 14:16:45: Encoding
Tuesday, 14:17:00: Encoding
Tuesday, 14:17:18: Encoding
Tuesday, 14:17:36: Encoding
Tuesday, 14:17:54: Encoding
Tuesday, 14:18:11: Encoding
Tuesday, 14:18:29: Encoding
Tuesday, 14:18:47: Encoding
Tuesday, 14:19:04: Encoding
Tuesday, 14:19:22: Encoding
Data rate 4.79 Mbps (4.66 video, 0.13 audio)
Tuesday, 14:19:25: Movie completed

14:19:25
14:16:30
--------
00:02:55

Interesting results. That's twice as fast as I would have expected given the results of my single 1.8GHz system (half would be 5:51 and then 11% off of that (to account for the higher clock speed) would be 5:12. I wonder if there are other factors other than core count (looks as if 2 maybe 2 1/2 cores is all this application can use) and clock speed.