latest posts

Never thought the day would come - but I have switched back to Intel for my primary desktop. My now almost three year old AMD FX-8350 and 990FX motherboard has been slowing dying on me for the last couple months. First a year ago the motherboard "lost" its ability to support 4x8gb DDR3 sticks providing me a much needed 32gb of ram. Then a few months later, I had to underclock the remaining 2 sticks of DDR to 1066mhz otherwise I would get BSODs throughout the day. Now for the last 3-4 months I have had to run a large house fan directly into my Corsair Carbide 540 - which has made VOIP meetings interesting when I have had to present. Long and short I was set to buy a brand new AMD FX-9590 and the newly released MSI 970 board that had USB 3.1 support, but it got me thinking - why am I paying for what is largely three year old technology? Knowing that when Zen is released next year I would be swapping whatever I chose to upgrade to, I decided I might as well go back to Intel for the next year while I wait for Zen to arrive.

Fortunately, Skylake has just arrived along with a slew of new motherboards. Seeing the prices of the Intel boards being considerably higher priced than AMD (even 990FX boards), I went middle of the ground with the Asus ROG Maximus VIII Hero which uses the Z170 chipset. For CPU, I was pleasantly surprised to find the newly released top end CPU from Intel, the 6700K was in my "AMD Pricepoint".

Three new features that I was excited to finally get was DDR4 support, USB 3.1 and having an M.2 slot for a future Samsung M.2 based SSD.

In addition, I was extremely curious how my FX-8350 and the i7 would compare. I know the i7 being built on a 14nm process versus the 32nm FX-8350, the i7 would produce less heat and use less power. My particular i7 uses 95W versus my FX-8350 which uses 125W, a reduction of 24%. Funny enough both are clocked at 4ghz and then there is still the i7 "Quad Core with Hyper-threading" versus the 4 module "8 core" of the FX-8350 (8 Integer Units and 4 Floating Point Units).


In addition to the i7 6700k "Skylake" processor and Asus ROG Maximus VIII Hero, I purchased a Corsair H100i GTX Extreme Performance Water cooler and 16gb of Corsair DDR4 3000. For the remainder of the system I am going to reuse my relatively new MSI Radeon R9 280X, hard drives, Corsair Carbide 540 and Corsair 750W power supply.


After having used Corsair closed loop coolers, it was pretty painless install in my Corsair Carbide 540. I mounted it in the front, which I am considering moving to the top and having air blow out the front (something to play with down the road). After installing the motherboard I was very surprised the overall layout has not changed much from the now 3 year old Asus Sabertooth 990FX motherboard I was using previously.


My first goto benchmark for comparing the two cpus was 3D Mark. Running with the same video card, I was hoping for a huge boost in CPU, fortunately I was correct.

Cloud Gate saw a 31% increase, Fire Strike 8%, Sky Diver 17% and Ice Storm 32%. For a CPU clocked at the same speed (4ghz) and half the cores (see above) to get between 8 and 32 percent increase was quite huge.

Next up was my other goto jcBENCH.

My earlier hunch about the integer cores looks to have been correct. With Hyper-Threading, I still only have 4 Integer Units while my old FX-8350 had 8. Interestingly enough - my FX-8350 actually did slightly better in Integer processing when using 8 threads. Floating Point was a completely different story - over double the performance in 8 threads - the IPC differences really show their colors there.

Closing Thoughts

Was it the large leap in performance I was expecting? Yes and no - in floating point operations (i.e. gaming/compute/etc), I can definitely see how things would perform much faster when all the cores are used. However in integer based tasks, I am not sure I will really see the advantage. In the coming weeks I will do some more comparisons and dive into some overclocking with the new machine. Still crossing my fingers for AMD to come through with Zen next year...
One machine that I had always wanted finally came to a pricepoint I saw as a good deal. This machine as expected from this blog's title, a Sun X4600 M2. This machine originally was released in Summer 2008 with the version I have in Late 2008. Luckily the version I have has 8 AMD Opteron 8384 Quad Core CPUs for a total of 32 cores. Most configurations I had been seeing on eBay are either Dual Core or lowerend Quad Core models.

Sun Sunfire X4600 M2 - Front

Sun Sunfire X4600 M2 - Inside

Sun Sunfire X4600 M2 - Hard Drives

I had the Mushkin 120gb and Western Digital Black hard drives laying around from a NewEgg sale a while back. I should note the SAS Controller on the X4600 is SATA II only, so your SATA III drives will be limited to SATA II's 300mb/sec.

One word of caution, during the initial bootup of the system the decibal level get quite extreme. After 10 seconds or so the noise gets to a bearable level, but no where near "living room" safe levels. As a result, I'm building a little rack in the garage since it stays around 50' all year round and can just use a Powerline adapter to keep it connected to my other machines.

Having purchased a HP Blade thinking I might one day own an HP Blade enclosure, I had 32gb of compatible DDR2 ECC ram laying around. One thing to note, the arrangement of ram is incredibly picky. If you're populating more than 4 DIMMs per CPU module they all need to match. I had at first just used the 16 DIMMs to populate the CPU modules as I found empty slots. Upon turning on the X4600 I was presented with only a fraction of the CPU modules and 16gb of ram (down from the 80gb it should have been). I ended up populating two CPU modules with the 16 2gb sticks I had from the HP Blade and then maxing out another with the original ram.

Sun Sunfire X4600 M2 - Extra ram from 
HP Blade

Operating System wise I chose Windows Server 2012 R2 since I had already started utilizing Hyper-V on my NAS for FreeBSD, OpenSUSE, Windows XP and Solaris VMs. To get Windows Server 2012 R2 installed I had to use an external USB drive and reduce the ECC Ram Settings in the bios to Good. Otherwise after cycling through the onboard SAS controller it would hang.

After applying all of the Windows Updates since 2012 R2 hit RTM, I pulled up Task Manager:

Sun Sunfire X4600 M2 - Windows Server 2012 R2 

Task Manager

Immediately afterwards, I pulled open jcBENCH to see how it compared to my other systems, not surprisingly it is the current leader in the world:

[bash] jcBENCH 0.9.850.0531(x86/Win32 Edition) (C) 2012-2014 Jarred Capellman Usage: jcBench [Number of Objects] [Number of Threads] Example: jcBench 100000 4 This would process 100000 objects with 4 threads Invalid or no arguments, using default benchmark of 150000 objects using 32 CPUS CPU Information --------------------- Manufacturer: AuthenticAMD Model: Quad-Core AMD Opteron(tm) Processor 8384 Count: 32x2693.27mhz Architecture: x86 --------------------- Running Benchmark.... Integer: 1897 Floating Point: 2265 [/bash] Pretty impressive for a 6 year old machine I believe, a part of me wishes I was still doing 3D Animation and Visual Effects on a routine basis as this would have made a rendering back in the day so much better with 3ds max and After Effects. Going forward this will be my primary Hyper-V server.

Anyone else who’s been around long enough to remember buying an AMD Thunderbird, Palomino, Barton etc. feeling the urge to build a new AMD system to replace their existing system?

My primary desktop is still being powered by a FX-8350 I bought November 2012 - while for everything I do (mainly Visual Studio and SQL Server) coupled with my Samsung SSDs, Radeons etc. it's ok, it just feels extremely weird to not have an upgrade path (ignoring the FX-9000 series) after years of doing a new year build every year or every other year (back in Athlon XP days it was every couple months). In the mean time I've built 2 Kabini systems this year, one as a ClearOS firewall and another as a HTPC and while those have worked out extremely well, there is a void in that my primary machine is relatively unchanged for almost 2 years.

I know there is always the Intel route - an i7 would be a definite upgrade, I know it is silly, I just can't bring myself to doing it. When I switched to AMD December 2000 with a 1 GHz Athlon it seemed like more of an experiment. AMD wasn’t new to making CPUs, but it was new in beating Intel in both price and performance. I had an aging Celeron 300A and tons of new First Person Shooters to play, the Athlon made sense then and if I had to do it again, I most definitely would. The summer before, in 1999 after having worked on the Army base doing Sys Admin work I had over a $1000 to burn – I almost pulled the trigger on building the parts for a Dual Pentium 3 600mz (buying only 1 CPU that summer since they were several hundred dollars). Thankfully, my dad convinced me to instead buy a 19” iiyama CRT (1600x1200) to replace my 15” CTX CRT (1024x768) – much more bang for the buck. Thus the following Christmas in 2000 I received some money and fortunately had enough to buy an Abit KT7-A, 256mb of PC133 ram and a 1ghz AMD Athlon (Socket A) – looking back had I still had my summer job in Germany, I might have built 2 of them since they were so cheap in comparison to the Pentium 3 idea the summer before.

I hope the new AM3+ platform replacement (latest rumor I read was next year) can really bring AMD back to the top. The FM2+ platform is nice, but the lack of 8 core variants without an iGPU is I feel definitely hurting AMD especially those sitting on the AM3+ platform like myself who would love to stay with AMD, but have nothing to upgrade to.

Just a quick update, I just wrapped up the 0.9.850.0531 release of jcBENCH for Mac OS X/x86. Previously there was a much older port going back to January 2012, so it is nice to finally have it running on the same codebase as the other platforms now. So please if you've got an Apple product, please go run jcBENCH as I only have an older iMac to compare results with.

On a somewhat funny note the iMac I do all my Xamarin work on is several factors slower than my $60 AMD Athlon 5350 I use in my firewall - funny how technology catches up at a fraction of the cost.
Over the weekend I picked up an AMD Athlon 5150 APU (4x1.6ghz) along with a MSI AM1I motherboard from Frys on sale for $59. A month or two ago I purchased the 5350 APU (4x2ghz) and an ASUS AM1-I, which had been working great since I set it up so I was curious how the 5150 performed along with the MSI motherboard.

MSI AM1I and AMD 5150

Big differences between the two is the inclusion of a Mini-PCLe slot (for a WiFi or SSD card) along with a PCIe 2.0 x16 slot (x4 mechanical). MSI AM1I and AMD 5150

AMD 5150

For those that have followed my blog for a while, I swapped the motherboard into my Lian Li PC-Q03B case that I bought back in 2012. In the time since I setup that machine, I had installed XBMC to stream TV, Movies and Music from my NAS (to be written up at a later date). Over time it became apparent the low powered AMD C-60 I had in there wasn't truly up to the task of streaming higher bit-rate 1080p video.

One thing I wasn't a big fan of after booting up the machine was MSI's choice of bios in comparison to my ASUS motherboards:

BIOS aside, I had no issues with the motherboard, Windows 8.1 updated itself without the need for a reinstall and I was ready to benchmark it.

Interestingly enough comparing jcBENCH the 5150 is roughly 40% slower in both Integer and Floating Point than my 5350. I included a couple other systems for comparisons:

AMD 5150 Benchmark Results

Saturday morning after resolving my Realtek RTL8111-GR Issues on my new ClearOS box, I ran into yet another error:


Knowing the AM1 Platform that my AMD Athlon 5350 APU/ASUS AM1I-A motherboard run on more than likely does not support IOMMU like my desktop's 990FX does I figured it was a detection issue with the Linux kernel that ClearOS 6.5 utlizes.

Doing some research into the issue there are a couple adjustments to your GRUB configuration that may or may not resolve the issue. In my case adjusting my GRUB arguments upon boot to include iommu=soft resolved the issue. I'm hoping down the road with newer Linux kernels the detection (if that even is the issue) gets better, but for those running an AMD "Kabini" APU and ran into this issue you'll at least be able to boot into your Linux distribution without any issues.

As some maybe aware, I recently purchased an Asus AM1I-A for a new ClearOS machine to run as a firewall. The installation for ClearOS 6 went extremely smoothly, but upon restarting I kept receiving kernel panic errors from eth1 (the onboard Realtek RTL8111-GR). After doing some investigating, it turns out RHEL and thereby ClearOS have an issue with loading the r8169 kernel module when it detects the RTL8111 (and the 8111 variants).

Sure enough after doing an lspci -k:
[bash] 02:00.0 Ethernet controller: Realtek Semiconductor Co., Ltd. RTL8111/8168/8411 PCI Express Gigabit Ethernet Controller (rev 11) Subsystem: ASUSTeK Computer Inc. Device 859e Kernel driver in use: r8169 Kernel modules: r8169 [/bash] The dreadful r8169 kernel module is the only module installed and in use. Thankfully you can download the r8168 x64-rpm here or wget

After downloading, simply run:
[bash] wget -i kmod-r8168-8.037.00-2.clearos.x86_64.rpm [/bash] and then:
[bash] modprobe r8168 [/bash] Then add Blacklist r8169 to the /etc/modprobe.d/anything.conf and then restart your machine.

Once your machine is backup, you can verify the correct r8168 module is loaded by re-running lspci -k:
[bash] 02:00.0 Ethernet controller: Realtek Semiconductor Co., Ltd. RTL8111/8168/8411 PCI Express Gigabit Ethernet Controller (rev 11) Subsystem: ASUSTeK Computer Inc. Device 859e Kernel driver in use: r8168 Kernel modules: r8168, r8169 [/bash] After installing and using the r8168 module I no longer received kernel panic errors and was able to utilize the onboard RTL8111-GR without issue. Hopefully this helps someone else who ran into the same issue I did.

I received my AUS AM1I-A and AMD 5350 (Kabini) APU along with the rest of the components for my new firewall system last night (a detailed build write up will come tomorrow night). For those curious about the raw CPU Integer and Floating Point performance here are the results from my newly updated Linux port of jcBENCH.

[bash] jcBENCH 0.8.755.0504(x86/Linux Edition) (C) 2012-2014 Jarred Capellman CPU Information --------------------- Manufacturer: AuthenticAMD Model: AMD Athlon(tm) 5350 APU with Radeon(tm) R3 Count: 4x2046.077mhz Architecture: x86 --------------------- Running Benchmark.... Integer: 1.99601 seconds Floating Point: 8.44405 seconds [/bash] In comparison to my FX-8350, with a 4 threaded limit, the Kabini based APU is actually faster than my FX-8350 (3.02 and 11.67 seconds respectively). In single and dual threaded tests, I imagine the results are far different. More thorough results will come tomorrow.

Over the Labor Day weekend, I happened to be in Silicon Valley on a short vacation and Fry's Electronics luckily had an Asus X55U Notebook on sale for $258. I had been looking for a long battery life laptop that wouldn't kill me if it ever got hurt during traveling (versus my much more expensive HP laptop). On top of that, I had wanted a Linux laptop to do cross-platform testing of jcDB, jcPIMP and the Mode Xngine. Planning ahead a bit, I brought an older Corsair Force 3 90gb SSD that I was no longer using and a Philips screw driver (yes folks, screw drivers are allowed by the TSA, it just can't be more than a few inches long).

Asus X55U Notebook

Asus X55U Notebook box contents

Asus X55U OpenSUSE Boot Menu

Specifications wise the laptop has:
-15.6" 1366x768 Screen (not the greatest quality, but definitely a lot better than expected at the price point)
-AMD E2-1800 (Dual 1.7ghz APU that clocks itself down to 2x850mhz when performance isn't needed)
-4gb DDR3 1066 (1 slot, upgradeable to 8gb)
-Radeon HD 7340 (DX11, 512mb of system ram is used)
-500gb Hitachi SATA Drive
-1 USB 2.0 and 1 USB 3.0 Port
-HDMI Output
-Gigabit Ethernet
-802.11n WiFi
-VGA Output
-Mic and Headphone Jack
-DVD-RW Drive
-SD/SDXC/SDHC Memory Card Slot

I should note, this APU does support AMD's Virtualization, so you can run Hyper-V, Xen, VMware Workstation etc. on this notebook. Coupled with the 8gb of ram support, this could be a decent portable VM notebook for the price.

Fortunately, doing the swap of the hard drive was extremely easy as opposed to some laptops that require taking apart the entire laptop (looking back at the Dell Inspiron I purchased in October 2010). Just 2 screws to pull off the back, which also contains the single DDR3 SO-DIMM slot.

Asus X55U Notebook (Bottom)

Corsair Force 3 90gb SSD

Curious if the system supported 8gb of DDR3 ram (since the manual didn't specify), I bought an 8gb DDR3-1333 Corsair SO-DIMM:

Corsair 8gb DDR3-1333 SO-DIMM

Swapped out the Hynix 4gb with the Corsair:

Asus X55U with the Corsair 8gb DDR3 SO-DIMM installed

And sure enough, the notebook supports 8gb:

Asus X55U with 8gb showing in the BIOS

While in the BIOS I should mention the charge when off feature this notebook has. Meaning with the lid closed, you can still charge your phone or media player. I wish my HP had that functionality.

Asus X55U BIOS Options

OpenSUSE 12.3 installed without a hitch (even the WiFi worked off the bat). After getting the system configured, the first thought I had was take the recent ia64/Linux port of jcBench and port it over to x86/Linux. On the flight back from SFO, I ported it over and thankfully it was only a re-compile with a slight tweak to the CPU detection.

How does the system perform?
[bash] jcBENCH 0.6.522.0928(x86/Linux Edition) (C) 2012-2013 Jarred Capellman CPU Information --------------------- Manufacturer: AuthenticAMD Model: AMD E2-1800 APU with Radeon(tm) HD Graphics Count: 2x1700.000mhz Architecture: x86/x86-64 --------------------- Running Benchmark.... Integer: 65.4932 seconds Floating Point: 35.6109 seconds [/bash] In comparison to my Silicon Graphics Prism (2x1.5ghz Itanium 2) it performs a little slower in Integer operations, but is nearly 3X faster in Floating Point Operations. In comparison to my HP DV7 laptop (AMD A10), it performs in the same dual threaded applications about 2X as slow, as expected with the slower clock rate and much smaller cache.

Overall, the notebook does exactly what I want and more for a $258 device. Build quality exceeds the Acer netbook and Dell Inspiron I had several years ago, coming close to my HP DV7, if only this Asus used the higher grade plastics. For those curious, battery life is about 4 hours with WiFi enabled the middle of the road screen brightness.
After getting my Silicon Graphics Prism ready to go on Saturday, and then getting the Intel C++ Compiler up and running on Sunday, my first task: getting jcBENCH ported over.

Not having done Linux development in a while (IRIX is similar, but different), I had to go through a few hurdles. One of which being, remembering the compiler macro for detecting Linux versus Win32 and IRIX. Luckily Sourceforge has a great listing of all the macros. After pushing through a few timespec_t issues (Linux's version is timespec), I was able to get it ported over while leaving in my #ifdef to retain portability.

A couple notes for me to remember and in case some else has the same issues:
*To fix the "undefined reference to 'clock_gettime' and 'clock_settime'"
-include "-l rt" to your g++ argument list

So how does my Dual Intel Itanium 2 1.3ghz/3mb L3 Cache do?

[shell] jcBENCH 0.6.522.0826(ia64/Linux Edition) (C) 2012-2013 Jarred Capellman CPU Information --------------------- Manufacturer: GenuineIntel Model: Itanium 2 Count: 2x1300.000000mhz Architecture: IA-64 --------------------- Running Benchmark.... Integer: 63.0576 seconds Floating Point: 119.499 seconds [/shell] In comparison to my other machines, Integer performance with both CPUs is some where between a single CPU of an A6-3500 (3x2.1ghz) and an A10-4600M (4x2.3ghz).

In comparison, Floating Pointing performance with both CPUs faired much better in that it was some where between my old Phenom II X2 545 (2x3ghz) and an A6-3500 (3x2.1ghz).

You can download the initial ia64/linux release here.

What's next? Now that I have some baseline metrics, tomorrow night I will swap in the spare 1.6ghz/9mb L2 Cache Itanium 2 CPUs I purchased a while back for my Silicon Graphics Altix 350 systems. Stay tuned for more details.
Curious if an external graphics card would work in the PCIe x16 Slot (X4 mechanically) on the ASUS C60M1-I I picked up a few weeks ago, I tried out a 1GB Gigabyte Radeon 6450. Gigabyte 6450 Radeon Gigabyte Radeon 6450 Card Sadly, the card would not post with the ASUS C60M1-I motherboard. I had tried reseating it, but to no avail. The card does in fact work as I put it back in an ASUS F1MA55 system and worked fine again. Not to be discouraged, I picked up an XFX Radeon 6670 HD to see if maybe there was some incompatibility with the 6450 and the motherboard. XFX Radeon 6670 XFX Radeon 6670 Card XFX Radeon 6670 installed into a Lian Li PC-Q06 Case Sure enough the card booted up just fine, the Radeon 6290 that is embedded in the C-60 is disabled when an external graphics card is used in case you were considering CrossFire or EyeFinity usage. With the several projects I am working on, I've only had time to do one benchmark, using 3D Mark 2011, here are the applicable scores for both the XFX 6670 and the 6290 in the C-60 itself:
3D Mark 2011 Scores for the AMD C-60's embedded Radeon 6290AMD C-60 APU (6290 Radeon) 3D Mark 2011 scores for the XFX Radeon 6670 with the AMD C-60AMD C-60 APU with an XFX 6670 Radeon
As you would expect, the Radeon gave a pretty sizable boost in performance. In real world testing, I tried out StarCraft II and was able to play at 720p with little lag unless there was considerable amount of Units on the screen at once (more than likely CPU Bound at that point). Going forward I will be using this box for OpenCL performance testing, so more to come for sure, but it's safe to say if you don't have a need to use the PCIe slot on the Asus C60M1-I, you're best bet if your intentions are gaming, is to get a decent Radeon 66XX and enjoy the benefits of offloading as much as you can to the GPU. As more and more applications rely on the GPU, a low watt CPU and higher powered GPU I imagine will become much more valued, but I could be way off in that prediction.
This morning I finally retired my AMD Phenom II X6 1090T CPU from my primary desktop. I had been using it since April 30th 2010, right when it first came out. Looking back, it's interesting to think the power that $309 bought back then and the $185 on the FX-8350 brings today. Just from a numerical standpoint, 6x3.2ghz (19.2ghz) versus 8x4ghz (32ghz) is mind blowing in my opinion. 12 years ago nearly to the day I was about to buy my first 1ghz AMD Athlon "Thunderbird" Socket A CPU. What is also interesting is that 2.5 years later AMD is still using AM3/AM3+, which for a consumer is great. Knowing with a simply bios update I can run the latest CPUs is a great to know. In my case, doing a bios update on my ASUS M5A99X EVO to get support for the just recently released Vishera series of FX CPUs from AMD. [caption id="attachment_1639" align="aligncenter" width="300"] AMD FX-8350 Tin[/caption] [caption id="attachment_1641" align="aligncenter" width="300"] AMD FX-8350 installed into my ASUS M5A99X[/caption] [caption id="attachment_1642" align="aligncenter" width="169"] AMD FX-8350 installed into my ASUS M5A99X[/caption] After installation with no surprise, the FX-8350 showed up properly and automatically increased my memory speed to 1866mhz (previously with my Phenom II the max available was 1600mhz). [caption id="attachment_1643" align="aligncenter" width="300"] AMD FX-8350 showing in the UEFI bios of my ASUS M5A99X[/caption] [caption id="attachment_1644" align="aligncenter" width="300"] AMD FX-8350 Detailed Info showing in the UEFI bios of my ASUS M5A99X[/caption] CPU-Z: [caption id="attachment_1645" align="aligncenter" width="300"] AMD FX-8350 in CPU-Z[/caption] And now the most interesting aspect of any upgrade. Can I justify the cost of the upgrade, especially when applications hadn't seemed sluggish. Integer Benchmark Results: [caption id="attachment_1647" align="aligncenter" width="300"] jcBENCH Integer Benchmarks[/caption] Floating Point Benchmark Results: [caption id="attachment_1648" align="aligncenter" width="300"] jcBENCH Floating Point Benchmark[/caption] I included a few extra CPUs recently benchmarked for comparison. First thoughts, Integer performance over the Phenom II X6 is over 200% across the board for single to 8 core applications/games, meaning the FX-8350 can do what the Phenom II X6 did with half the CPUs leaving the other half for other tasks or making multi-threaded tasks 200% faster theoretically. This is also shown in the A10-4655M CPU, in 4 threads, my laptop was actually faster than my desktop as far as integer only work is concerned. Kudos to AMD for making such a dramatic difference in integer performance. Floating Point results were a bit more interesting. Having seen quite a bit drop off in comparison to the Integer results, I was curious if the FX-8350 would hit the same hurdles. Sure enough because of the drop off of the 1 to 1 relationship between Integer Cores and Floating Point Cores in the Phenom II architecture in favor of a 2 to 1 ratio in the latest generations of AMD's CPUs, the Phenom II actually beat out the much higher clocked FX-8350, albeit the more threads, the less of an impact it made. Definitely more benchmarks will ensue with real world tests of Visual Studio 2012 compiling and After Effects CS6 rendering. Stay tuned.
I just received my ASUS C60M1-I today and figured with the lack of information on this board on the internet I'd post my findings. [caption id="attachment_1602" align="aligncenter" width="300"] ASUS C60M1-I Box Contents[/caption] In the box you'll get the manual, DVD-ROM with drivers, Powered By ASUS sticker, 2 18" SATA III cables and the I/O Shield. Note the I/O Shield is not like the higher end ASUS boards with the padding. [caption id="attachment_1596" align="aligncenter" width="300"] ASUS C60M1-I Motherboard[/caption] [caption id="attachment_1595" align="aligncenter" width="300"] ASUS C60M1-I Ports[/caption] [caption id="attachment_1597" align="aligncenter" width="134"] ASUS C60M1-I DDR3 RAM Slots[/caption] [caption id="attachment_1598" align="aligncenter" width="300"] ASUS C60M1-I 6 SATA III 6gb ports[/caption] First off I was curious if the 8GB maximum was really true. Having 4 8gb Corsair Vengeance DDR3-1833 sticks awaiting to be put into my primary desktop, I popped a pair into the board. Sure enough the motherboard read them without a hitch: [caption id="attachment_1603" align="aligncenter" width="269"] Corsair Vengeance 16GB DDR3-1833[/caption] [caption id="attachment_1604" align="aligncenter" width="300"] Corsair Vengeance 16GB DDR3-1833 Stick[/caption] [caption id="attachment_1605" align="aligncenter" width="300"] ASUS C60M1-I BIOS showing 16GB DDR3[/caption] Also curious is the option of running the ram at 1333mhz. Using 1600mhz and 1833mhz DDR3, you could in theory run it at 1333mhz and keep the timings really nice. [caption id="attachment_1606" align="aligncenter" width="300"] ASUS C60M1-I BIOS showing 1333mhz Option[/caption] Next up was seeing if the board supported RAID of any type, this turned out to be false. A little more investigation on the motherboard itself, the South Bridge is the FCH A50M. It also does not support USB 3.0 nor has a native Gigabit controller. In this case ASUS went with a RealTek 8111F Gigabit Controller. I personally have had the 8111e which ran fine and assume this is just a revision of that controller. If someone has more info on it, please let me know. The big points for me were Jumbo Frame support to 9k and Gigabit, both of which are features of the 8111f that is on this board. Another thing to note is the lack of HDMI port. Not a huge deal with readily available DVI->HDMI adapters, but something to consider if that is make or break. The more expensive/faster ASUS E45M1-I DELUXE has an HDMI port (in addition to USB 3.0), though it does draw more power than the C60M1-I. Another thing I was curious about was total power draw of the system. Based on what I read the CPU is 9W and the A50M Southbridge uses 4.7W, coupled with 2 4gb Sticks I expected maybe 20W. The total idle usage (sitting in the bios) is 23W. The total usage under 100% load is 39W. I should note this was done with Antec VP-450, a higher efficiency power supply might bring that number down a bit. [caption id="attachment_1613" align="aligncenter" width="300"] Antec VP-450 450W Power Supply[/caption] So there you have it, pretty low wattage. For comparison, my Acer Aspire One AO522-BZ465 that I got in June 2011 uses 24W idle in bios and 40W under load. The last question I had, something I imagine a lot of people would be wondering for NAS purposes is if the PCIe x16 slot (x4 mechanical) would support non-graphics cards. I had a 240GB OCZ Revodrive x4 card that I was hoping to use in this board so I gave it a shot, sure enough it worked without any hassle. [caption id="attachment_1607" align="aligncenter" width="300"] 240GB OCZ Revodrive PCIe SSD[/caption] [caption id="attachment_1608" align="aligncenter" width="300"] 240GB OCZ Revodrive PCIe SSD card[/caption] [caption id="attachment_1609" align="aligncenter" width="300"] OCZ Revodrive PCIe SSD BIOS[/caption] I don't have any other PCIe controllers laying around to test, but I imagine you would be ok with them. Options I would consider, maybe Infiniband for a MOSIX SSI or a SAS PCIe x4 controller? Onto the more fun stuff, the benchmarks. [caption id="attachment_1616" align="aligncenter" width="300"] jcBENCH Integer Comparison[/caption] [caption id="attachment_1617" align="aligncenter" width="300"] jcBENCH Floating Point Comparison[/caption] [caption id="attachment_1618" align="aligncenter" width="300"] Windows Experience Index Comparison[/caption] The numbers can speak for themselves, but I should point out the vastly better CPU performance in jcBENCH over the C-50 CPU. The Turbo clock speed of the C-60 really does make a huge difference. So to summarize:
  1. ~23W of usage at idle (BIOS)
  2. ~40W of usage at full power
  3. 16gb of DDR3 is the max this board will take, not 8gb as mentioned on the ASUS Website
  4. Ram can be set to run at 1333mhz not just 1066mhz like on the website
  5. No HDMI Port
  6. No USB 3.0 Controller
  7. No RAID Controller like that found in the 7xx/8xx/9xx AMD Desktop Chipsets
  8. The PCIe x16 (x4 mechanical) can be used for non-graphics cards
  9. Turbo mode of the C-60 does give a considerable boost in CPU performance
  10. OpenCL 1.1 Support
Any comments, suggestions, wanting more information, please let me know.
Luckily, there was a HP Pavilion dmz1 E-350 (2x1.6ghz) at work that a co-worker was able to run jcBench on so if you were curious about Ontario vs. Zacate vs. Champlain vs. Trinity core pure CPU performance now is your chance to see. Integer Performance: [caption id="attachment_1285" align="aligncenter" width="300"] AMD Ontario vs Zacate vs Trinity vs Champlain CPU Integer Performance[/caption] Floating Point Performance: [caption id="attachment_1286" align="aligncenter" width="300"] AMD Ontario vs Zacate vs Trinity vs Champlain CPU Floating Point Performance[/caption] As expected the E-350 was in between the 2010 Champlain CPU, but what was more interesting was the lack of a dedicated FPU for each Integer Core in the Trinity CPU in the Floating Point benchmark. While still faster in the A10 versus the P920, it wasn't as dramatic as the Integer performance difference. It should be interesting how AMD tries to combat the floating point deficit it most likely faces against Intel's cpus. If I had to guess, I'd say more emphasis on using the integrated Radeon 7xxx series is key to the success of Trinity. Regardless of the platform using OpenCL or C++ AMP is a better route to go nowadays than strictly programming for a CPU as a traditional multi-threaded development would go. Definitely will revisit this in the coming weeks when the new version of jcBench that supports TPL, C++ AMP and OpenCL is released.
Not that I give much creedence to the Windows Experience Index, but here is my score with the addition of the 16GB of Corsair DDR3-1600 and SanDisk Extreme 240GB SSD: [caption id="attachment_1273" align="aligncenter" width="300"] Windows Experience Index for my HP DV7-7010US[/caption] What is interesting in those numbers is that my older Phenom II P920 (4x1.6ghz) got rated just under that in CPU results. After using the laptop for a week now, it definitely feels and responds orders faster than my older Dell. So I turned to jcBench, a much clearer picture (I included my C-50 Netbook for comparison): Integer Results (note lower is better, the legend indicates the number of threads tested) [caption id="attachment_1274" align="aligncenter" width="300"] AMD C-50 vs AMD Phenom II P920 vs AMD A10-4600M in Integer Performance[/caption] Looking at these numbers, the results make a lot more sense. Neither the Phenom II nor the C-50 have a "Turbo" mode in which the unused cores get tuned down and the used cores get ramped up. For instance the A10 in my DV7 will ramp up to 3.2ghz from 2.3ghz. Thus the interesting result in 2 threads of the Phenom II to 1 thread on the A10, nearly equal time (2x1.6ghz vs 1x3.2ghz effectively). I will run Floating Point on my P920 tonight and update the post, but I feel it will be fairly similar to the Integer results. However, the shared FPU on the Trinity CPUs should make it more interesting since the Phenom II had 1 FPU per core.
After nearly 2 years, I'm finally retiring my Dell M5010 (AMD Phenom II P920/1GB Radeon 4650), it was very reliable up until a few months ago, but for the price (~$700) back in September 2010 I had expected more, especially in the build quality department. Fast forward to now, I had been waiting for the new line of APUs from AMD to come out, codenamed "Trinity" for quite some time, skipping over the llano series and hoping for a nice notebook for under $1000. The HP DV7 thus far is delivering on that. With a 1600x900 17.3" screen, Island style (not the annoying to type on "Chicklet"), pretty nice sounding speakers/subwoofer (Beats Audio) and the A10-4600/7660G Radeon it's pretty speedy. The build quality is also much higher than I was even expecting, it's not an EliteBook level, but pretty close. [caption id="attachment_1247" align="aligncenter" width="300" caption="HP DV7-7010US Box"][/caption] [caption id="attachment_1249" align="aligncenter" width="300" caption="HP DV7-7010US Metal Looking Finish"][/caption] [caption id="attachment_1254" align="aligncenter" width="300" caption="HP DV7-7010US running Windows 8"][/caption] [caption id="attachment_1239" align="aligncenter" width="300" caption="HP DV7-7010US Keyboard and Touchpad"][/caption] Before even turning it on for the first time I swapped out the 6GB of DDR3 for 16GB (2x8GB) Corsair Vengeance DDR3-1600 ram and a 240GB Sandisk Extreme SSD. Something that had been annoying me to death on my Acer AO522 AMD C-50 Netbook and my Dell Laptop was the fact I had basically take apart the whole machine to upgrade the ram or hard disk. I am happy to say, one screw off the bottom plastic plate and a total of 6 screws for the harddrive you can accomplish the entire swap. [caption id="attachment_1241" align="aligncenter" width="300" caption="16GB of Corsair DDR3-1600 Vengeance RAM"][/caption] [caption id="attachment_1252" align="aligncenter" width="300" caption="SanDisk Extreme 240GB SSD"][/caption] [caption id="attachment_1245" align="aligncenter" width="300" caption="HP DV7-7010US Bottom Panel Removed"][/caption] I installed the Windows 8 x64 Release Preview and Visual Studio 2012 RC on it thus far, still looking for Windows 8 drivers for the 7660G...
I've been working on an OpenCL WPF version of jcBench for the last 2 or 3 weeks in between life, thus the lack of posting. However this morning at a nice time of 6:30 AM, I found a new tool to assist in OpenCL development, the AMD APP Kernel Analyzer. In my code for testing I've just been setting my OpenCL kernel program to a string and then sending it to the ComputeProgram function like so:
string clKerenel = @"__kernel void computePythag(global long * num, global double * thirdSide) {
     size_t index = get_global_id(0); for (long x = 2; x < num[index] + 2; x++) {
     for (long y = x; y < num[index] + 2; y++) {
     double length = sqrt((double)(x * x) + (double)(y * y)); length *= length; thirdSide[index] = length; }
return; }
"; ComputeProgram program = new ComputeProgram(_cContext, clProgramSource); program.Build(null, null, null, IntPtr.Zero); ]]>
This is problematic as any error isn't captured until runtime best case scenario or you just get a BSOD (as I did last night). This is where AMD's KernelAnalyzer comes into play. [caption id="attachment_1224" align="aligncenter" width="300" caption="AMD APP Kernel Analyzer"][/caption] Note the syntax highlighting and if there were any errors during compilation the Compiler Output window like in any other development tool. An extra feature that I just realized how useful it really is, is that the ability to target to difference generations/platforms of AMD GPUs. I knew the R7XX (4000 series Radeon HD) only had OpenCL 1.0 support, but I didn't realize (naively thinking) that the same program wouldn't compile cleanly across the board. Luckily I still have 2 R7XX series GPUs in use (one in my laptop and another in my secondary desktop), but interesting nonetheless. Definitely more to come on the OpenCL front tonight...
Got some more results add to the comparison: [caption id="attachment_1218" align="aligncenter" width="300" caption="jcBench Expanded Integer Results - 5/15/2012"][/caption] [caption id="attachment_1219" align="aligncenter" width="300" caption="jcBench Expanded Floating Point Results - 5/15/2012"][/caption] Tonight I am hoping to add a quick comparison page so anyone could ask, well how does a Dual 600mhz Octane perform against a 600mhz O2?
Below are some interesting results of the floating point performance differences between MIPS and AMD cpus. [caption id="attachment_1105" align="aligncenter" width="300" caption="jcBench Floating Point Performance"][/caption] The biggest thing to note, is the effect Level 2 cache has on Floating Point performance. The 4mb Level 2 cache in the R16000, clearly helps to compensate for the massive difference in clock speed. Nearly a 1 to 1 relationship between a the 6x3.2ghz Phenom II and the 4x800 MIPS R16k. So bottom line, Level 2 cache makes up for megahertz almost by a factor of 4 in these cases. It's a shame the fastest MIPS R16000 only ran at 1ghz and is extremely rare. More benchmarking later this week...
I was able to add several more machines to the comparison with the help of a friend over at Nekochan. [caption id="attachment_1102" align="aligncenter" width="300" caption="jcBench Integer Comparison 2"][/caption] Very interesting how MIPS scales and how much of a difference 100mhz and double the Level 2 cache makes an effect on speed.
Just finished getting the features for jcBench 0.2 completed. The big addition is the separate test of integer and floating point numbers. The reason for the addition of this test is that I heard years ago that the size of Level 2 cache directly affected performance of Floating Point operations. You would always hear of the RISC cpus having several MegaBytes of cache, while my first 1ghz Athlon (Thunderbird), December 2000 only had 256kb. As I get older, I get more and more scrupulous over things I hear now or had heard in the past thus the need for me to prove to myself one or the other. I'm still working on going back and re-running the floating point tests so that will come later today, but here are the integer performance results. Note the y-axis is the number of seconds taken to complete the test, so lower is better. [caption id="attachment_1097" align="aligncenter" width="300" caption="jcBench 0.2 integer comparison"][/caption] Kind of a wide range of CPUs, ranging from a netbook cpu in the C-50, to a mobile cpu in the P920 to desktops cpus. The differences based on my current findings vary much more greatly with floating point operations. A key things I got from this data:
  1. Single Threaded, across the board was ridiculously slow, even with AMD's Turbo Core technology that ramps up a core or two and slows down the unused cores. Another unsettling fact for developers that continue to not write parallel programs.
  2. The biggest jump was from 1 thread to 2 threads across the board
  3. MIPS R14000A 600mhz CPU is slightly faster than a C-50 in both single and 2 threaded tests. Finally found a very near equal comparison, I'm wondering with the Turbo Core on the C-60 if it brings it inline.
  4. numalink really does scale, even over the now not defined as "fast" numalink 3 connection, scaling it across 2 Origin 300s using all 8 cpus really did increase performance (44 seconds versus 12 seconds).
More to come later today with floating point results...
Just got the initial C port of jcBench completed. Right now there are IRIX 6.5 MIPS IV and Win32 x86 binaries working, I'm hoping to add additional functionality and then merge back in the changes I made to the original 4 platforms. I should note the performance numbers between the 2 will not be comparable. I rewrote the actual benchmarking algorithm to be solely integer based, that's not to say I won't add a floating point, but it made sense after porting the C# code to C. That being said, after finding out a while back on how Task Parallel Library (TPL) really works, my implementation of multi-threading using POSIX, does things a little differently.

Where the TPL starts off with one thread and as it continues processing increases the threads dynamically, my implementation simply takes the number of threads specified via the command line, divides the work (in my case the number of objects) by the number of threads and kicks off the threads from the start. While TPLs implementation is great for work that you don't know if it will really even hit the maximum number of cpus/cores efficiently, for my case it actually hinders performance. I'm now wondering if you can specify from the start how many threads to kick off? If not, Microsoft, maybe add support for that? I've got a couple scenarios I know for instance would benefit from at least 4-8 threads initially, especially for data migration that I prefer to do in C# versus SSIS (call me a control freak).

Back to jcBench, at least with the current algorithm, it appears that a MIPS 600mhz R14000A with 4MB of L2 cache is roughly equivalent to a 1200mhz Phenom II with 512kb L2 cache and 6mb of L3 cache at least in Integer performance. This is based on a couple runs of the new version of jcBench. It'll be interesting to see with numalink if it continues this to 1 to 2 ratio. I'm hoping to see how different generations of AMD cpus compare to the now 10 year old MIPS cpu.
Finally got a Silicon Graphics Visual Workstation 330. Not a traditional Silicon Graphics by any means, but the case was neat and fufills a nostalgic purpose. [caption id="attachment_29" align="aligncenter" width="550" caption="SGI VS 330 in box"][/caption] [caption id="attachment_31" align="aligncenter" width="550" caption="SGI VS 330 Sideview"][/caption] Not wanting to be limited to Dual Pentium 3 933mhz CPUs and no SATA support, I used a spare CPU, ram and motherboard I had laying around. [caption id="attachment_32" align="aligncenter" width="550" caption="Asus M4A89GTD/USB3, Corsair XMS3 16gb, Athlon II X3 435 and Crucial 64gb"][/caption] In addition I swapped out the rear fan, being 11 years old it was quite noisy. I had a spare CoolerMaster 120mm low-rpm Blue LED fan that fit perfectly in the back. All in all, pretty smooth swap. Downloaded the Network Install CD of OpenSuse 12.1, installed it and was ready to go in under an hour. In it's completed form: [caption id="attachment_33" align="aligncenter" width="550" caption="Completed swap of parts out of the SGI VS 330"][/caption] All that's left now is to de-scuff the case to bring it back closer to a new state.
Just benchmarked my netbook, web server, desktop and a few machines at work. Some interesting results, to really take advantage of x number of cores you need operations that are complex and or in a longer duration. Looking at a 24 Core Opteron server run the same number of tasks was actually slightly slower than my 6 core Phenom II. However, running a task that is 10X longer, the 24 core runs 3X faster. Factoring in that the Opteron cores are 900mhz or 29% slower, that's 4X faster, directly equating to the fact the Opteron had 4X as many cores. More details as I benchmark more machines.
Been reading: Definitely recommended reading when you get a chance. Learned a ton about Wireless Security. If you're on Fios definitely make sure the Verizon Router they give you is on WPA at least, by default they come with WEP turned on and basically broadcast the password inside the packets it sends out in broadcast. On a related note, if you want to mess with your router, checkout this site. I flashed my router to it a month or so ago. The web interface is so much smoother than the factory firmware. Going along with the wireless theme, I got 2 netbooks yesterday, well ordered them at least. Got a pair of these bad boys: Got 2 spare Corsair 4gb DDR3 so-dimms, an OCZ Vertex 30GB SDD and an OCZ Vertex 2 60gb SSD. I plan to swap out the 1gb sticks and 250gb 5400 drives in it with those parts, should bring the performance up considerably. I plan to do extensively benchmarking before and after so everyone will know how each benefited and what percentage. I definitely will be wiping the Windows 7 Starter Edition for OpenSuse. They should be both come Tuesday, latest Wednesday so stay tuned!
I recently got a Rackable Systems 2U Dual Dual Core Opteron system off eBay for dirt cheap.  The problem however though as I noted previously was the noise, Antec .  Got 3 Antec fans and had my Antec 500w ATX power supply connected externally, much quieter, but I didn't feel comfortable leaving my ESXi server opened like that so I scrambled.  Found my original 4U file server case I purchased back in 2009 and started the migration. [caption id="attachment_4001" align="aligncenter" width="300" caption="Old 4U File Server case on the left, 2U Rackable Systems Case on the right"][/caption]


[caption id="attachment_4002" align="aligncenter" width="225" caption="Motherboard inside 4U File Server Case"][/caption]


[caption id="attachment_4003" align="aligncenter" width="225" caption="Transplant powered on"][/caption]


[caption id="attachment_4004" align="aligncenter" width="300" caption="Revised Server Stack"][/caption]


Just scored 2 brand new AMD Opteron 2360SEs off eBay for $69.99 shipped.  These are the highest clocked Quad Cores out there at 2.5ghz.  I had a Dual Quad Opteron 2344HE back in Spring 2009, albeit they were 1.7ghz cores for a total of 13.6ghz versus my new server at 20ghz. It is kind of interesting thinking about it now, my main rig has a 6 core Phenom II has 19.2ghz of power.  It might have been cheaper to simply get another 1090T, 890FX/990FX series AMD motherboard, 16gb of ram and throw it in the same case? Let's see...
  • Phenom II 1090T (3.2ghzX6) - $190
  • Asus Sabertooth 990FX motherboard - $210
  • 16gb G.Skill DDR3-1600 - $180
For a total of $580 Versuses...
  • Rackable Systems with 2U Case, 2X2214HE Opterons - $150
  • 2 Opteron 2360SEs - $140
  • 10gb Kingston DDR2-5300 ECC - $90
For a total of $380, plus I plan on selling the original 2U case and Opterons so probably will end up spending $320. $200 off the bat plus whatever I can get back.  In addition I have plenty of room for more ram 16 slots versus 4 slots on the 990FX motherboard.  Worth it?  I think so  
Call me crazy, but I simply couldn't stand having an Intel running in the house.  I had bought a Dual Core Atom (D510) CPU/Motherboard a few years ago, originally to run Mac OS X on.  It bounced around between my NAS, HTPC, Windows Media Center 7 Extender and finally a Firewall/Gateway a month or so ago running ClearOS. Over the last weekend I realized I had a new AMD Sempron 140 still sealed, checked prices on the new 5 series Asus motherboards, saw it was only $60 on Amazon and picked up a 2gb G.skill DDR3-1333 stick.  As fate would have it, NewEgg sent over a Norco RPC-270 2U case.  Took about an hour but I got everything installed and up and running again on the new AMD platform.  Below are the build pictures. [caption id="attachment_3981" align="aligncenter" width="225" caption="Empty Norco RPC-270 Case"][/caption] [caption id="attachment_3982" align="aligncenter" width="225" caption="Asus M5A78L AM3+ Motherboard Installed"][/caption] [caption id="attachment_3983" align="aligncenter" width="225" caption="Everything Installed"][/caption] [caption id="attachment_3984" align="aligncenter" width="300" caption="Gotta love AMD CPUs, unlocked the 2nd core without a blink of an eye"][/caption] [caption id="attachment_3985" align="aligncenter" width="300" caption="The stack continues to get bigger"][/caption]