Video encoding i7 2.8 vs. i7 3.4

[yustas]

When it comes to video encoding and DVD creation using iMovie, how much of a difference there is between 21" i7 2.8 vs. 27" i7 3.4? I know that 3.4 i7 is faster, but what about the video card, does it make a difference?

Also, with only 512 MB video memory on 21" i7 2.8, is it enough video memory to run Windows 7 via Parallels?

 

[iMikeT]

Video encoding is a processor intensive task. You should notice approximately a 21% increase in performance from the 3.4GHz CPU versus the 2.8GHz.

 

[yustas]

Video encoding is a processor intensive task. You should notice approximately a 21% increase in performance from the 3.4GHz CPU versus the 2.8GHz.

That sounds like a lot. Where have you seen this benchmark?

 

[mrfoof82]

That sounds like a lot. Where have you seen this benchmark?

Video encoding doesn't really need a benchmark across different CPUs clock speeds for the same architecture. It scales almost perfectly linearly with CPU performance. Transactional database performance pretty much follows the same rule -- it scales almost perfectly linearly with clock speed. 10% faster clock = 10% lower processing times.

This is almost always the case with any "embarrassingly parallel" task where the data can be easily partitioned prior to work, and video encoding is unquestionably under that umbrella.

 

[pawtracks]

Here are some numbers to look at Link

I have a 2009 27" 2.8GHz i7 and a 2011 27" 3.4GHz i7 SSD.
Not exactly what you are asking but maybe I will do a side by side handbrake comparison and see how they stack up.

 

[yustas]

What about 512 MB of video memory? I do not play games, but what are the limitations of having 512 MB vs. 1 GB in terms of using Photoshop, iMovie, running Parallels and using an additional monitor?

 

[yustas]

Here are some numbers to look at Link

I have a 2009 27" 2.8GHz i7 and a 2011 27" 3.4GHz i7 SSD.
Not exactly what you are asking but maybe I will do a side by side handbrake comparison and see how they stack up.

Thanks you! Exactly what I was looking for, but I am not sure I understand some of the numbers.

For example, Speedmark 6.5 individual application test results: Parallels World-Bench Multitask test:

21.5-inch iMac 2.8GHz Core i7 (BTO): 254
27-inch iMac 3.4GHz Core i7 (BTO): 226

It says: "All other test results in the above chart are in seconds; lower results are better," - okay, so in this test 3.4 i7 is faster, but faster doing what in Parallels? 28 seconds is not a lot, unless we are talking about opening apps, of course. Is it 28 seconds faster opening apps or...?

 

[rupnok]

One thing to keep in mind with all those benchmarks, is they are short burst. Meaning, you get to see performance under peak Turbo. I am way more interested in benchmarks for long encoding/sustained usage. As heat builds, particularly on the low TDP 2.8 i7, I would assume it wouldn't be able to Turbo up as much.

I can't believe that the 2.2 Ghz i7 in the MBP is on par with the 3.1 i5, other than the benchmarks are point-in-time, and not indicative of ongoing use/heat build up.

 

[pawtracks]

Thanks you! Exactly what I was looking for, but I am not sure I understand some of the numbers.

For example, Speedmark 6.5 individual application test results: Parallels World-Bench Multitask test:

21.5-inch iMac 2.8GHz Core i7 (BTO): 254
27-inch iMac 3.4GHz Core i7 (BTO): 226

It says: "All other test results in the above chart are in seconds; lower results are better," - okay, so in this test 3.4 i7 is faster, but faster doing what in Parallels? 28 seconds is not a lot, unless we are talking about opening apps, of course. Is it 28 seconds faster opening apps or...?

World-Bench is a common benchmark to compare systems. Sounds like they ran a specific world bench test or series of tests using a parallels VM (virtual machine) since it only runs on a PC (OS), not a Mac (OS). and yes, the 3.4 i7 did the "test" 28 seconds faster than the 2.8 i7, 11% faster is 11% faster, in that particular test.

 

[yustas]

World-Bench is a common benchmark to compare systems. Sounds like they ran a specific world bench test or series of tests using a parallels VM (virtual machine) since it only runs on a PC (OS), not a Mac (OS). and yes, the 3.4 i7 did the "test" 28 seconds faster than the 2.8 i7, 11% faster is 11% faster, in that particular test.

But what does this test do?

 

[pawtracks]

But what does this test do?

Tests a system to see how fast it can perform certain tasks, then you have a benchmark for comparison. Not a simple answer, "this is what it does", it depends on what you tell it to do and what programs are installed, what you want to benchmark, what you intend to use, etc...

I wouldn't worry too much about the details, just understand how to read the scores to see which is superior to the other.

Wiki