2012-07-30

Just update about my activities (NetBSD, CPU collection)

There are no new blog posts for some time now. This time however I actually don't have anything to write. There are no hardware changes recently. Though I plan to increase total RAM to 16GB (4x4GB) an make my system to use quad channel memory instead of dual channel one next week. NetBSD developers didn't find fast answer why integrated NIC doesn't work (on ASRock X79 Extreme4-M). I wasn't successful too despite the fact all other major BSDs (Free and Open) works ok with it. Because of this I also plan to insert some chip 1GBit PCI-E network card (most likely D-Link DGE-560T with marvell NIC). I hope that it will be a temporary solution until solution to enable correct operation of integrated one will be found. I also changed NVIDIA GeForce GTS450 graphics card to fanless ATI Radeon 2400XT. The main reason was a good NetBSD support but downside was higher power usage at idle (yes, GTS450 ate noticeably less energy on idle). Finally, there are problems with USB devices in NetBSD, I always need to reconnect mouse or keyboard after system is booted...

NetBSD reached 6.0 Beta 2 stage at current point and heading towards first release candidate. When we can expect release I still can't even predict. If I remember correctly  5.1 was released about 6 months after first RC release... Nevertheless you can download updated 5.1.2 version or use 6 daily branch. What is more, VT-310DP second CPU bug is finally  fixed on version 6 branch. It was VIA Eden CPU specific problem that it hid cmpxchg8b instruction support because of some Windows NT(!) bug. The worst thing that there was no BIOS entry to change that too (I saw such option on some Jetway VIA based motherboards). Fortunately, you can expose support for this instruction using software methods (some assembly). Despite this fix there are still some ACPI problems with VIA chipsets, because of this you still need to use -2 parameter (disable ACPI) at boot. There is a patch to this problem too and the system will boot correctly with ACPI enabled but you will see a constant interrupt storm in CPU activity. Because of this patch is not very suitable for production use. Maybe something will change in distant future :).

Finally, I updated my CPU collection photo album. It is difficult for me to list all new items as they came at different time but there are some new interesting ones. I bought new camera (Olympus SP-610UZ) so you can expect a little bit better photos in the future. Some of them I have already refreshed. The main new CPUs include Integraph Clipper C300 (new architecture and rare, hard to find item), HP PA-RISC 8800 (dual core, Itanium socket), HP Alpha 21364 1300 MHz (if wikipedia is not wrong it is the fastest Alpha), Zilog Z380 32-bit MPU, set of 6 Kendall Square Research supercomputer chips, Infineon MIPS CPU (new company in the list). So that is all for now.

2012-04-08

Sandy Bridge-EP platform

While Intel's Ivy Bridge CPUs are on the horizon I happened to get Sandy Bridge-E sample on my hands. It is Sandy Bridge-EP 4S more precisely. This CPU will become a member of Xeon E5-4600 series when they will be released and they are primarily intended for 4-way servers. My sample is packed into 2011-land FC-LGA package and works on LGA 2011 socket.  Despite its intended purpose to be a server CPU it can be used on desktop perfectly too. My sample's s-spec is QAFF and its specifications are as follows:
  •  3GHz frequency (30x multiplier), max 3.3 GHz on turbo boost
  • 10 MB L3 cache
  • 4 cores, 8 threads
  • 32 nm 
  • 130W TDP (not confirmed)
  • support for AVX, SSE4.2 and other modern instructions and features like in most sandy bridge cpus.
So I would call it "low-end" or "mainstream" CPU on targeted platform and it is comparable to Core i7 3820 desktop CPU by specs but has much less frequency (3820 clocked at 3.6 GHz by default). CPU multiplier is locked. 

There is always a risk if you buy engineering sample CPU. Firstly, no motherboard will support it officially and you need to find at least some information on supported ones (mostly you need to know at least one supported motherboard from specific manufacturer and then compare their motherboard specifications, BIOS and CPU support to make decision if you are not satisfied this the one you found as compatible). In the worst case you probably will be able to sell incompatible motherboard for some little loss (of course, if you don't live in the small country like me). Secondly, you won't have any warranty. If CPU will die, nobody will replace it. Finally, you can't write a lot about it until official CPUs are released (Intel doesn't like leaked CPUs and can take actions on sellers and on even on bloggers). Nevertheless this CPU has more or less the same core as other Sandy Bridge-E(P) CPUs so you can always imagine performance by reading articles on similar items. When you consider all these factors and still decide to buy a CPU you should be ready to lost the money in the worst case. However, it is still aren't very risky decision in my experience. CPU is probably one of the most reliable parts compared to other computer parts. If you spend enough time to check motherboard compatibility and have enough sources to prove your decision this move becomes even less risky. On the bright side ES CPU can be purchased for less money than production chip (not applicable for first samples of new architecture or platform or very rare collectible old items).

Personally it is my second engineering sample CPU (used for actual computer, not for collection purposes). First one was Xeon E5540 ES Q1E2 Nehalem CPU. Unfortunately, X58 platform wasn't very satisfactory for my purposes and I replaced it with Athlon II X2 235E on AMD 780G platform. It was great platform until Jetway JNC81 motherboard died. Jetway replaced it with JNC84-E but it wasn't so friendly with NetBSD. What is more, I wanted more performance for compiling, folding and other tasks. Though, I still didn't plan to change computer initially in the near future but I bought Sandy Bridge-EP CPU at a reasonable price and decided build the platform on it.

I wanted to go as cheaply as possible because of this I bought relatively cheap ASRock Extreme4-M motherboard, Thermaltake Contac39 CPU cooler and reused some parts from retired mother's Phenom II X2 555 computer (like RAM and graphics card) and my computer (HDD). I plan to add two more RAM modules to make them work on 4 channel mode but currently system works on 2 channel mode. I also would like to get less powerful and more power efficient graphics card (GeForce GTS450 now) but it depends on how successfully I will sell my parts on second hand market. Nevertheless the platform still cost quite a lot as X79 motherboards are pricey and even the cheapest one cost about double of what I spent on one motherboard typically. Extreme-4M wasn't the cheapest one but Intel boards didn't have enough sources to prove compatibility (ironically) and Extreme-4M was more appealing for me than Extreme-3 which one was the cheapest solution from ASRock line.

So my current setup is:
System setup went without big problems. CPU was recognized as GenuineIntel 3GHz. It works on 1.2GHz on idle. Motherboard uses modern UEFI BIOS. I needed to set 1600 MHz for RAM manually as they worked as 1333MHz by default. No other big changes were needed. Fedora 16 booted without visible problems and worked perfectly. NetBSD wasn't so good. It booted better than on JNC84-E platform (it typically didn't load on first try and needed repeated boot command in the bootloader. This problem disappeared by surprise for me), but NIC (Broadcom BCM57781) didn't work correctly (as result no internet and network), GTS450 graphics card wasn't supported too (X.org failed to start). Keyboard needed reconnect on the first boot but problem didn't reoccur on other times.  So currently I need to solve these issues somehow. On the other hand I expected more worse situation as platform was relatively new. These two problems are quite small and solvable. Of course, USB3.0 is not supported on NetBSD but I don't have any devices anyway. The whole platform uses 86-92W on idle and ~160W on full CPU load (by folding@home). For comparison unlocked 4 core Phenom II X2 555 used around 200W and more with the same graphics card and RAM.

QAFF ES 3.0 GHz

2012-02-29

EPIA-M900 power usage and performance

This is the last article from my series about my EPIA-M900 system (first, second). This time I'll focus on power usage numbers, some small performance tests and overall feel.

I used  Tunex multifunction power meter to measure a power usage. All tests were made using Windows 7 64-bit edition. The whole system (including eH1 graphics card and HDD) without monitor uses 44-46W on idle. 44W is more constant than higher values. I tested system on BIOS without graphics card too and it showed 29-31W values. So graphics card adds about 15W on idle. Later I used AIDA64 stress tests and GPU-Z render test to simulate system's maximum power usage with and without HDD stress test. On full load (CPU and GPU) power meter showed 75-77W and 73-74W with and without HDD stress test respectively.  That was the maximum power usage numbers I managed to get on this system. GPU render test raised power usage to 55-56W. CPU load only increased power usage more than GPU load to 68-69W. Finally, while playing Starcraft II this range was between 70-71W. So it is difficult to say how much power usage increased because of GPU or CPU as other components takes some too and overall power usage doesn't increase on the same level as on CPU or GPU load separately.

While power usage numbers are quite good we should look into performance. As of 26 February, 2010 1.6GHz (1.73GHz on turbo) is the maximum frequency you can get from dual core VIA CPUs. VIA Nano X2 L4350E TDP is 27.5W. So you can compare it to low usage mobile CPU (if I am correct, current standard mobile CPU's TDP goes from 25W-45W typically) by TDP. Unfortunately, performance is on the lower range and you need to compare them to Intel Atom and AMD APU from bobcat core series. This is the area where VIA CPU is really competitive. But this article won't make any summary and won't make big comparisons . I will provide only some raw numbers for your investigation and reference. 


Firstly I used y-cruncher multithreaded pi benchmark application. I used 50,000,000 decimal digits test and Linux 0.5.5 Build 9187 version. Total result was 126.278 seconds for this test. Single core VIA Nano L2200@1.6GHz completed test in 299.358 seconds. U2300@997MHz needed 497.043 seconds. The same test was slower on Windows for unknown for me reason (about 20 seconds in total). If you want to compare this result to other CPUs, it is comparable to Intel Celeron T3100 result.

I also ran AIDA64 tests (
v2.00.1700). I didn't tested L2200 but AIDA64 had results in their table. Memory read using DDR3-1066 SDRAM 7-7-7-20 CR2 was 3747 MB/s (Nano L2200 with DDR2 3352  MB/s). While memory write test was much faster than L2200, 7138 MB/s against 3157 MB/s. Memory copy 4944 MB/s vs 2759 MB/s. Memory latency was the same 117.3 ns on both systems. CPU Queen - 5890 vs 2580. This test was a little bit faster than AMD E-350 (5169). CPU PhotoWorxx - 5083 vs 2493. For reference Celeron 420 had almost the same result, E-350 had quite lower result at 4375. CPU Zlib - 34.2 MB/s, L2200 - 14.7 MB/s, E-350 31.0 MB/s, Core i5-650 HT - 101.1 MB/s. Padlock module helps VIA Nano CPUs to shine on CPU AES test where L4350E got 86803 points and lost only to CPUs with Intel AES support (though considerably) but others without it lost considerably (even 12xOpteron 2431 got 78761 points), L2200 - 40002 points. CPU Hash test gained 1054MB/s vs 549MB/s (E-350 achieved 326MB/s only). FPU was always quite a weak place for Centaur CPUs but Nano architecture improved FPU performance. FPU V8 436 vs 318, This test was bad for VIA CPUs, even E-350 got a little bit more here. On the hand it showed better results at FPU Julia and Mandel tests 1862vs892 and 832vs428 respectively (E-350 got results in between closer to L2200). Finally FPU SinJulia was a weak place too: 302vs132 while E-350 - 506. This is all tests I made. Tests showed quite mixed results but I believe that L4350 is comparable to first generation dual core AMD K8 CPUs or last generation Pentium 4 CPUs. Somewhere they will be quite faster, elsewhere much slower. They are faster than dual core HT Intel Atom CPUs (at least first generation). The last thing I could test was folding@home project. Unfortunately, this CPU is not suitable for SMP client. It is just to slow (dual core phenom ii@3.2GHz will do 1% 12-15 minutes than this CPU will need 2.5 hours). I also added Windows 7 performance index last time it is here once again (the CPU got minimum 3.9 result):

Windows 7 performance index
In general, it feels that CPU is much slower than Phenom II X2. Active processes like antivirus, flash, program loading grabs quite much from CPU. It is expected result of course. You can't compare apples and oranges this time. However, if you want to feel comfortable at your desktop, this system can be a little to slow. Though it is not that bad after you get into it. I want to change HDD to SSD some day. This should solve slower loading times and I guess should help to feel more responsive and fast while using the system. The only program which used hardware GPU acceleration in videos was Windows Media Explorer (MPC classic unfortunately took only CPU). I will try to investigate this in the future.

I didn't do any GPU tests. I tried to run some DX10/DX11 benchmarks but eH1 was unable to run them at normal FPS. However I played Starcraft II at 1280x1024 without any noticeable problems and with good graphics settings. While playing GPU temperature didn't raised more than 6-7°C from starting point (never reached 50°C).  I believe you can play most modern games with such configuration and not very big resolutions without problems.


So this is it about this system. I hope you enjoyed these articles and found them helpful. Thank you for reading.

2012-02-16

EPIA-M900 system setup and Windows 7 start up

After I tested some open source operating systems I've decided not to play to long and started to assemble the main configuration for the desktop PC. It consists of EPIA-M900 motherboard, eH-1 graphics card, 2xKingston KVR1333D3S9/4G RAM (I am not really that Kingston fan but lately all RAM comes from them and I can't complain, they have very good compatibility with most boards), Samsung HD161HJ 160GB hard drive and In Win BM-639 case. Initially I planned to add CD-ROM too but some modifications disallowed that (later on that). RAM works at 1066MHz effective speed because it is the maximum VX900 supports.

The main purpose was to replace that annoying small CPU fan. So I perforated one side of the case above the CPU/chipset heatsink and installed 120mm fan. This "project" went ok and I was able to remove default fan. CPU cores temperature goes from 43°C on idle to 63-66°C max on load (speedfan show 5 degrees less than AIDA64 at the same time). CPU temperature is ~55°C by AIDA64. System temperature never rises above 31°C. The downside of this solution was that the fan blocked ability to add CD-ROM as original case design placed construction for 5.25'' drive just a little bit below that side of the case. GPU temperature range starts from 41.3°C on idle to 46-48°C on load. I firstly made one mistake during assembling because one wire stuck between GPU heatsink preventing fan from rotating. This lead to temperature from 81°C on idle to more than 126°C on load. You can understand from this that eH1 can work without fan :). Graphics card fan is surprisingly silent. I've read a lot of angry comments about active cooling solution in this card but it seems that they aren't reasonable this time (I know that fans tends to fail after some time but on the other hand you need to keep good airflow with passive cooling solution and need fans somewhere anyway). Unfortunately, the PCI-E riser was only 1.1 compatible so I needed to turn on the switch on graphics card to support slower PCI-E version. I tested that motherboard supports 2.0 version actually (at least graphics card worked switch in off mode). Finally, PCI-E slot is x16 physically only. x8 is effective speed of that slot.


AIDA64 CPUID
The other task was to fasten graphics card because I bought not suitable riser (I needed right angle one but I bought straight one. It lead to the problem that riser raised graphics card much above case panel for cards bracket fastening). The solution was simple. I inserted graphics card upside down (fan is on the top now). I needed to cut the bracket a little bit and make one hole in the it and the case for the screw but this solution needed minimal effort and almost no sacrifice for the future use of the case or graphics card (yes, the original bracket will not be suitable for normal insertion any more but I have replacements and you probably will find them at various IT stores too). Unfortunately my camera is broken now because of that I couldn't to make photos of these modifications but I'll try to do them in the future.


GPU-Z (some information inaccurate)
The last thing I wanted to do before Windows installation was BIOS update. I found the newer version on viaembedded web page and I made an upgrade. Firstly, it seemed that everything went ok. However, after I loaded system defaults, motherboard just didn't boot anymore. It was very unpleasant moment because I though that motherboard just died because of unsuitable BIOS version. Fortunately it booted after I changed RAM to some spares I had had at home. BIOS showed incorrect checksum error on boot but I was very happy nevertheless as motherboard was alive at least. I downgraded BIOS back to old version (fortunately I made a backup of it because you couldn't download it from viaembedded web page). I don't know why VIA decided to add this BIOS version without any release notes and description but they will gain lots of unsatisfied users with such irresponsible move. These days BIOS upgrades are simple and painless process usually, because of this users will try to do this from time to time. Such surprises should be avoided and BIOS purpose and suitability for particular model stated clearly. Anyway I recommend to use original BIOS versions with VIA motherboards :)...


After all preparations I was able to try Windows 7 (64-bit) finally. I wasn't installing them cleanly, just tried to boot into installed system from previous computer. This process went smoothly... almost. Every time I tried to connect to the network, Windows was hanging up and rebooting with good, old BSOD in the end. Solution was simple yet again, I needed to install network drivers from CD or web page. It seems that Windows update provided faulty drivers for some reason. I also installed the newest 500 series Chrome drivers from s3graphics web page instead of the ones from CD or viaembedded eH1 web page. All other devices were recognized by Windows and successfully installed into the system. Due to system resources requirement for physical memory addresses Windows sees 7.25GB of RAM instead of 8GB. I will discuss some performance results and power usage in the next article.
Windows 7 rating

2012-01-30

I've chose EPIA-M900 for my VIA Nano X2 setup

While VIA prepares some new motherboards like EPIA-M910 and EPIA-920 (last one will be equipped with quad core CPU and VX11 (VT3456) chipset which has integrated DX11 compatible GPU) I've made a decision on my own configuration. It is quite unusual and consists of VIA embedded parts mainly.  So I've decided to take EPIA-M900 (EPIA-M900-16L) motherboard and eH1 graphics card. I've already received this hardware and started to test it.

In Win BM-639, VIA eH1, VIA EPIA-M900 boxes

EPIA-M900 is the most quality made motherboard from VIA embedded I've ever seen (especially compared to the most of VB series and old models). However, I admit that some recent models are similar by quality too. On the other hand this quality comes at substantially higher price than I would like to see. Unfortunately cooling system is made in VIA "fashion". It consists of the big aluminum heatsink over CPU and chipset and small, annoyingly loud fan. On the bright side the fan speed is controlled by CPU load using smart fan function in BIOS. However, it still emits unpleasant sound even at lower speed. GPU has a fan too but it seems less audible (no speed control). EPIA-M900 box is visually not appealing as it looks like simple corrugated cardboard box (I've never saw the one like that for motherboards earlier but as it is targeted for embedded market VIA probably decided to save money on box visual appearance). You will find a SATA cable, driver's CD and I/O shield besides the board itself. Yet again it is quite ordinary minimal set of accessories you'll find in most VIA products. There was two main reasons I chose this board. Firstly, it was the only VIA Nano X2 board which had a PCI-E x16 (effective speed up to x8) slot, secondly EPIA-M900 had two RAM slots and supported 8GB RAM while VB8004 (the other candidate) only one. Besides, it was easier to get than VB8004 with its graphics module.

EPIA-M900

VIA eH1 has more visual appealing and more standard box for such products. Accessories include DVI-VGA adapter and driver's CD. There is one caveat that you'll find a Linux tux image on the box indicating this graphics card "support" Linux but you won't find any drivers neither on cd nor in VIA Embedded web page. You can download some Linux drivers on s3graphics web site but you still won't be able to use them on modern Linux distributions as it has some code that won't compile on current kernel. I guess the card itself is supported by those proprietary drivers but I couldn't test them due to reason I mentioned above. There are no open source drivers for this GPU so the only way to use this card is VESA driver or try an older distributions (with 2.6.38 kernel or maybe even less). What is more, it doesn't use solid state capacitors as S3 graphics similar products.

VIA eH1 graphics card
The motherboard itself had some surprises that I hadn't notice or they hadn't been clearly defined in specifications. Firstly, most pin headers was smaller in size than standard ones. USB headers are surprisingly standard ones but front panel audio, fan, kbms and COM headers are mini variants. I saw smaller fan header (but forgot to check that they were only ones) and I knew about smaller COM headers but they were unneeded for me. Unfortunately I didn't know that front panel audio header and kbms header would be mini ones too. I actually don't know where to get adapters for them right now too. Second surprise was a missing RAID support in BIOS firmware (or I didn't found it). I though that such RAID support existed in all boards currently and you even had no need to check that. 

I tried some open source operating systems during this weekend. Linux (Fedora 16, openSUSE 12.1) and FreeBSD installed ok. NetBSD didn't boot (I guess because it doesn't support VX900 chipset still) but dual core CPU was recognized (UPDATE 2012-05-02: VX900 IDE support was added recently, so NetBSD 6_BETA currently boots on this board). Haiku boot sequence ended in kdebug with some stack trace. PC-BSD didn't boot too but it seems that it was a DVD boot loader problem because FreeBSD installed without any problems and surprisingly recognized all motherboard's integrated hardware (as you may or may not know PC-BSD is just desktop orientated FreeBSD variant). As I mentioned earlier eH1 graphics card and its HD audio is not supported on any open source OSes (except VESA driver).

I'll try to make some tests in open source OSes and/or Windows later on but I don't make any promises this time. I hope to test power usage too. To end this article on good note the board itself feels really good. It uses the most modern chips from VIA like VT2021 HD audio, VT6130 Gigabit Ethernet controller, VX900 integrated graphics with H.264 decoding acceleration on Windows, 8GB support (unfortunately USB 3.0 was not added for some unknown reason as the place for two connectors and VIA Labs chip was designed on the board). Smart fan control makes default fan bearable on idle. Finally dual core CPU has modern instructions support up to SSE4.1 and Intel compatible virtualization and it feels quite fast (I compiled quite a lot of source during this weekend, y-cruncher result is comparable to Merom 3M dual core pentium or celeron CPUs). Finally, the board quality gives hopes that VIA will continue to make more boards like this in the future and even improve them. BIOS could me more capable but it is embedded board so requirements are minimal for them I guess. Quite good FreeBSD support was a pleasant surprise too. I can't tell much about eH1 GPU right now as it will show its capabilities on Windows only.

P.S. I've updated my last  article on VIA Nano X2 solutions several times too. Probably I'll keep to do this for some time in the future.