Click for Portable Storage Solutions!
The Source for Mac Performance News and Reviews
Review: Newer Tech's MAXpowr G4/400 ZIF
G4 CPU Upgrade for AIO/Beige/B&W G3s
|Benchmark tests were run with MacBench 5.0 as it is the accepted Mac standard. I've also included results with three memory bandwidth benchmarks to see what the G4 CPU offers in that area, and to check Apple's claims of 2x higher bandwidth in their Sawtooth G4 systems. As you'll see below - there were some very odd and interesting results from the memory bandwidth tests.
BYTEmark tests are not shown since it returned a 0 score on the Sawtooth G4/AGP system. I'm not a big fan of BYTEmark as it is not a good indicator of other important system performance areas like disk, memory and graphics card. BYTEmark did run on a G4 upgraded Genesis system and results were shown in my First Look at the XLR8 G4/400 upgrade back in August.
Remember the most important results are on the next page of the review which covers real world applications performance. This is where the rubber meets the road so to speak. Benchmarks have their place, but actual applications performance is what really matters. As you'll see in the Photoshop 5.5 test results, until benchmarks like MacBench take advantage of Altivec, they are not an indicator of what the G4 is potentially capable of. However, MacBench CPU scores seem to be a pretty accurate gauge of non-Altivec G4 performance except that the higher FPU MacBench scores don't seem to show any benefit over the G3 in applications.
Be aware that Macbench Graphics scores will vary depending on graphics mode and the installed video card. In Macbench 5.0 a 1000 score is the baseline based on performance with an Apple Beige G3/300 running millions colors, 1152x870, so consider this when evaluating any scores at lower resolutions and color depths.
Since the two Macs tested here have graphics cards that use the same Rage128 chip, performance is similar. In all but the Quake2 tests at 640x480, the 66MHz PCI slot card in the B&W G3 showed similar performance to the AGP slot version in the Sawtooth G4/450.
MacBench 5.0 Tests: The graph below compares the follow configurations and systems
All systems had OS 9, 256MB of RAM, VM Off, and OpenGL 1.1.2 with its ATI driver update. All G4 tests (Sawtooth and Upgrades) had the 4 OS 9 Altivec extensions active.
MacBench 5.0 Results
Notes: Some explanation of the MacBench graph and test components.
G4timedemo is a benchmark for showing the benefit of Altivec support with G4 CPUs. I ran tests with each of the system/upgrade combos and graphed the results. As you can see, a G4 CPU w/Altivec support does make a huge difference here (60% gain with a G4 CPU of the same speed). All tests were run with the desktop video mode of 1024x768, thousands colors and G4timedemo set to 'amazing' (max) quality.
G4timedemo results seem to vary a few FPS from run to run but you can clearly see the dramatic difference a G4 w/Altivec makes in performance. I repeated the tests several times (each after a clean reboot) and noted the best score of 3 runs. According to some of my previous test data, OS 8.6 reported faster results with this benchmark (as it has in many applications tests as well).
Memory Bandwidth Tests:
I want to caution readers to not be too concerned with the results below. Although there were some strange results in many cases in the memory bandwidth tests, rely more on the real world application performance on the next page of this review rather than the results of pure benchmarks like these.
I used 3 tools to test for memory performance. Memory Bench, Stream and Newer Tech's new Gauge Pro. In the images below you'll see results of a stock B&W G3/400, the same system with G4 CPU upgrades (at both 400 and 450MHz speeds) and a new Apple G4/450 AGP (Sawtooth) system. All systems were running OS 9 and had 256MB of PC100 RAM. The Apple G4/450 system had all 2-2-2 clock SDRAM. The B&W G3 had 3-2-2 RAM. Although I doubt it mattered, when a G4 CPU was present the 4 Altivec OS 9 extensions were enabled.
Since the Apple G4/450 AGP (Sawtooth) system uses the 'maxbus' (now officially called MPX bus) feature of the G4 that is not available on any previous motherboard design, I was curious to see what effect this would have on memory bandwidth performance. Michiro Isobe recently commented on MPX bus advantages in my G4 forum, noting that for a single processor system, the only benefit is the elimination of idle time between burst accesses to memory.
The efficiency of the L1 and L2 caches in CPUs are the reason why a 50MHz system bus speed older Mac with the same type and speed of CPU upgrade generally keeps pace with modern Macs that have 100MHz bus speeds in most applications tests.
It appears both XLR8 and Apple set some supervisor registers in the G4 CPU (for compatibility some say) that affect memory bandwidth. As you'll see in the applications tests page of this review, the differences seen in memory bandwidth tests don't equate into significant advantages in most common applications and games however. For comparison purposes, most but not all of the test info from my past XLR8 G4/400Z review is provided to show how their software (at least the versions up to 1.43b0) affects memory bandwidth. I retested with the latest 1.4.3b0 XLR8 control but saw no improvement in this particular area.
It is interesting to note that the Sawtooth shows dramatically higher main memory bandwidth than the B&W G3 that used the same PC100 RAM (same speed, type and bus width). Look at the Memory Bench results below with data sizes that exceed the backside cache size to see that Apple's claims of a 2x improvement over the previous G3 design seems to be true.
Memory Bench Results:
First I show comparison results of the B&W G3/400 before and after the G4/400 CPU Upgrade.
Note the massive increase in L1 cache speeds on data sizes up to 16K with the same speed G4 vs the G3 CPU (in some cases more than 3 times faster at the same clock speed):
As I noted in the previous review of that upgrade, tests without the XLR8 extension active resulted in 159.6MHz cache speeds rather than 200MHz, but showed higher memory test results due to the G4s registers not being modified. Tests with the latest 1.4.3b0 software showed no real change in this area. In real world apps tests like Infini-D rendering, performance was better with the XLR8 software enabled (400/200) vs 400/159.6 with no XLR8 software enabled. (Now you see why my reviews take so long to produce, as I run many tests and often document many variations.)
Now let's look at the Apple G4/450 Sawtooth which uses the G4's 'MaxBus' design. Note the Sawtooth's maxbus design has higher L2 cache and main memory transfer rates than the B&W G3's motherboard, even when running the same CPU, Cache and Bus speeds. However as shown on the Applications tests page, you won't see much advantage from this with the majority of today's applications. It's a welcome improvement however and applications that move a huge amount of data should realize some gain.
Apple's claims of 2x higher bandwidth with the Sawtooth appears to be true according to the results above. Since the bus width and speed is the same as the B&W G3, the G4/Sawtooth 'MPX' bus feature must be responsible for the higher IO rates.
Stream is one of the most popular and widely known memory tests. The images below show comparisons of the B&W G3/400 vs a G4/400 upgrade in the same system and an Apple Sawtooth Apple G4/450 AGP.
Now the Sawtooth G4/450 AGP system Stream results
The final test used the built-in memory bandwith test of Newer Tech's GaugePro utility. The results with the Sawtooth don't track the other memory bandwith benchmarks however, as GaugePro reports the Newer G4/400 has more than 20% higher memory speeds. I'm not sure this makes sense given the results from other benchmarks.
Newer Tech's Gauge Pro Results:
Newer Tech's latest Gauge Pro utility also includes a memory bandwidth test feature. (Gauge Pro should ship with their new G4 cards.) What's interesting is that Newer Tech's G4 software control test results show a B&W G3 with a 400MHz G4 CPU upgrade has significantly faster memory speeds than the Sawtooth G4/450 system. The same thing was noted in the XLR8 G4/400 review as long as the XLR8 software extension was not used. Enabling the XLR8 software dramatically reduced GaugePro's reported performance rates as you'll see below.
According to test software by Michiro Isobe, Apple and XLR8's software are changing values in the G4's supervisior registers that affect memory bandwidth. As noted in my XLR8 G4/400 review, this is said to be done for compability reasons. Michiro had a G4 booster utility that changed the registers to improve performance on a Sawtooth (to higher rates than noted here on any system) but commented it required a G4 2.7 stepping CPU for reliability. My Sawtooth and every G4 upgrade I've seen so far is a 2.6 stepping or earlier (the supplied Newer Tech G4 was a 2.2 stepping, although their shipping versions may be later revs since a G4/400 CPU slot upgrade they sent had a 2.6 rev G4).
Note G4 CPU Stepping (CPU Version) is 2.2
Now the same test without the XLR8 Extension Active:
Now Sawtooth G4/450 AGP System Results:
Again it appears the XLR8 G4 software and Apple Sawtooth G4 change the default G4 supervisor registers. XLR8 commented on my initial observations of lower memory bandwidth in the MAChSpeed G4/400z review by saying these changes were done to ensure compatibility and address errata in the 2.6 and earlier G4 steppings. From the Newer Tech results, it appears they are not changing these registers but so far I've not seen any negatives from their software using a G4 2.2 stepping CPU. I don't have some of hardware mentioned to test (Media 100 was one item as I remember). Retrospect backups with verification, a long time acid test of CPU upgrades, ran perfectly.
Michiro Isobe wrote a utility to restore these registers but it may not be reliable with CPUs earlier than the current 2.7 stepping (his Sawtooth is a 2.7 rev). The register changes were not 100% stable on my 2.7 stepping G4 Sawtooth. Booting from OS 8.6 and opening a 2nd drive with OS 9 stalled the system (hourglass cursor). Michiro said he did not see that on his 2.7 stepping G4 Sawtooth system. There were also some random errors in Quake2 when changing OpenGL video modes. So perhaps these register changes are there for a reason with 2.6 and earlier G4s, although as I mentioned I've not seen any problems so far with the Newer Tech G4 software in a B&W G3 review machine.
These 'pure' benchmarks are interesting, but not directly related to performance in most real world applications as shown on the next page of the review.
|Summary: In other than (most) memory bandwidth tests, G4 upgrades score about as well as Apple's latest Sawtooth G4 system. MacBench CPU scores are disappointing after all the G4 supercomputer hype, as a G3 CPU of the same speed consistently scores higher than the G4 by a bit. The G4's FPU scores are higher than the G3, but that doesn't seem to be reflected in most real world applications.
Benchmark performance is interesting for technical reasons, however I prefer to rely on real world application performance since that's what really counts. Application and 3D Game tests are covered in the next page of this review. Or you may use the links below to jump to a specific page.
- or -
Copyright © Mike, 1999.
Users of the web site must read and are bound by the terms and conditions of use.