Just checked, there's no Nexus 9 scores there. If you mean the "HTC Volantis" scores, those are clearly not the Nexus 9. (2.5GHz vs 2.3GHz, respectively)
Engadget had this to say about Geekbench scores on the Nexus 9:
" Curious, I pitted the K1-toting Nexus 9 and Shield tablet against each other in a few more tests -- the 9 boasted a stronger single-core score than the Shield in Geekbench 3 (1,643 vs. 1,074), but the multi-core score definitely skewed in the Shield's favor."
PCWorld had this to say about Geekbench scores on the Nexus 9:
"In GeekBench 3, the Nexus 9 scored 3358 in the multi-core test"
A lot of reviews are out already, its quite clear that the selling point is 64-bit Android 5. Which really won't get much use or benefit considering the mount of RAM.
This feels like a repeat of the Tegra 3 Surface RT.
The processor is listed as Flounder, and the model is named htc nexus 9. Most of the scores are in the mid 1900s for single-core, but there are several over 2k, and some under 1900. It should be here : http://browser.primatelabs.com/geekb...ge=3&q=nexus+9 For example: http://browser.primatelabs.com/geekb...aseline=672893
Geekbench frequency reporting is notoriously wrong on most platforms except for iOS.And as you can see, that SoC is rated at 2.5GHz. The Nexus 9, the real actual product sold, runs at 2.3GHz.
Geekbench frequency reporting is notoriously wrong on most platforms except for iOS.
Then it should not be used at all, for anything.
Tho I don't agree with you, everything else I have seen seems correct.
Got any info on that?
Geekbench frequency reporting is notoriously wrong on most platforms except for iOS.
I'm pretty sure it lists it at 2499 MHz. But either way for the iPad Air 2, the frequencies are changing too. Either way, that's the legit score no doubt.
Could you let me know of scenarios where we report the wrong frequency? I'm aware of an issue with big.LITTLE systems but otherwise Geekbench should report the correct frequency under Android.
I was responding to Nothingness.
That score is not on the "Android Benchmark Chart".
That score is of the countless "Nexus 9" scores running at 2.5GHz.
The Nexus 9 has a 2.3GHz cap.
There is only one iPad Air 2 on the "iOS Benchmark Chart".
That single-core score of almost a 90% performance higher than the Shield Tablet must be obvious, somewhere, somehow and it isn't, in any review.
That's the advantage of designing the OS, the browser, and the hardware.while you are here, how well does geekbench work for cross os comparisons? IOS always seems to get a higher score compared to even higher end android handset results?
while you are here, how well does geekbench work for cross os comparisons? IOS always seems to get a higher score compared to even higher end android handset results?
Can someone compare an iPhone 4 with an android phone with the same A9 CPU? AFAIK it was clocked at only 800MHz and Android phones with an A9 were clocked at 1GHz an upwards, still we can account for the clock difference. I bet that even at 800MHz the iPhone 4 will score higher than an android phone with the CPU clocked at 1GHz. That way we will know how much the OS influences the results.
That will not be really useful, for a15s, krait, denver, a57s, etc look nothing like an a9. When you have dedicate silicon for specific test such as encryption you can't really extrapolate very old architectures to new ones which have new designs.
Can someone compare an iPhone 4 with an android phone with the same A9 CPU? AFAIK it was clocked at only 800MHz and Android phones with an A9 were clocked at 1GHz an upwards, still we can account for the clock difference. I bet that even at 800MHz the iPhone 4 will score higher than an android phone with the CPU clocked at 1GHz. That way we will know how much the OS influences the results.
Sorry, I was indeed considering recent scores with b.L systems... Can you explain how you get frequency? It's interesting that Nexus 9 is reported at 2.5GHz instead of the marketed 2.3. Perhaps some form of turbo?Could you let me know of scenarios where we report the wrong frequency? I'm aware of an issue with big.LITTLE systems but otherwise Geekbench should report the correct frequency under Android.
The one I just posted was Mashable's review, plus phonearena had 1903.
Could you let me know of scenarios where we report the wrong frequency? I'm aware of an issue with big.LITTLE systems but otherwise Geekbench should report the correct frequency under Android.
Why is Geekbench showing an overclocked TK1 Denver on the Nexus 9? The product is sold at 2.3GHz.
The performance of a Nexus 9 varies as much as going from a Snapdragon 600 SoC up to a Snapdragon 805 SoC. How can you target an application to run well in such a mixed bag?
So that would explain why it is showing 2.5GHz in Geekbench 3, because the SoC is rated up to 2.5GHz, but the actual max operating frequency set by the driver in Nexus 9 is 2.3GHz.
That is not really true. In the majority of cases, the CPU performance of TK1-Denver is well above Snapdragon 800/801/805, including Geekbench 3 single-core, Kraken, Google Octane, Vellamo Metal, Vellamo Browser, DMIPS, SpecInt2k, SpecFP2k, 16GB Memcpy/Memread, etc. The only benchmarks it does poorly in is Sunspider and Basemark OS II "Web". Keep in mind that Sunspider performance is heavily dependent on browser optimizations too (the Tegra 4-powered Surface 2 has higher Sunspider performance than the A7-powered iPad Air!). Subjectively, in previews it was noted that web pages load and render just as fast on the dual-core Nexus 9 as they do on tri-core iPad Air 2, so clearly the poor Sunspider and Basemark OS II "Web" benchmark result is not indicative of the real world web browsing experience on Nexus 9.
Can someone compare an iPhone 4 with an android phone with the same A9 CPU? AFAIK it was clocked at only 800MHz and Android phones with an A9 were clocked at 1GHz an upwards, still we can account for the clock difference. I bet that even at 800MHz the iPhone 4 will score higher than an android phone with the CPU clocked at 1GHz. That way we will know how much the OS influences the results.
The RAM was perhaps not the same? Or the CPU revision was not the same? Also do you know that even when a benchmark doesn't make use of OS calls its performance will vary depending on the OS? even running on the same machine and OS can show several percent differences (as an example most SPEC scores are measured just after a reboot of the machine to get the best possible result). I remember running a computational program through Wine under Linux a few years ago: it was 10% faster than on Windows despite not making any significant use of the OS or the C library.You can look at GeekBench between the iPhone 4 and the original Samsung Galaxy S and it's variants as they both use the "Hummingbird" CPU. It's about as close to identical hardware as you can fine to compare an Apple product and an Android. Omitting the outlier results in general Geekbench shows the Galaxy about 5-10% faster even though it has 25% higher clock speed. So interpret that how you want. IMO Geekbench is more a software benchmark then a hardware as true hardware benchmark should show similar results regardless of platform.
For example, us two can have a Nexus 9, but mine scores 30+% higher than yours.
No, it doesn't work like that. Nexus 9 has different power saving modes, so obviously the performance can vary depending on what mode is selected. Most reviews/previews are showing pretty consistent performance in the default settings, and the margin of error is only ~ 10% for most CPU and GPU benchmarks (in Geekbench 3, the vast majority of scores are between 1800-2000 for single-core performance, and between 3200-3500 for multi-core performance).