Falconise

Close enough! 369 would do it, but you'd run 369 only if 370 was a problem I think. The math works best with numbers containing 3 (or a product of 3?), like 133, 166, 233, 266, 333, 533, 667, 800, 1066, etc. Usually the board will compensatre for this, but you will find common speeds where these numbers keep popping up. Another goal speed would be 3.56GHz, or 375 x 9.5. But like I said earlier, the next big step up from 333FSB is 400FSB, so a new goal of 3.8GHz is in order: 400 x 9.5 = 3800MHz or 3.8GHz rounded. Now that should be noticable, and hopefully achievable without having to increase any voltages! Beyond that, I think we'd need to first find out what the maximum FSB the board is capable of, it may not do much more than 400MHz. The CPU almost definately will, and certainly the board may too but I don't know this. The process of finding the max FSB is not difficult, uses the same settings we changed above but with different tactics. We'll get there later perhaps.

Manual StockSpeed Settings Sometimes, when you take too many things off Auto at once, the system fails to boot. Not a big problem, the P5E should boot into safe mode, let you enter the BIOS and make changes. If you get too far astray, simply load Profile 1 in the tools menu. Look at the screenshots on this page of the review I was reading: http://www.xbitlabs.com/articles/mainboards/display/asus-p5e_6.html We will set the system up exactly like that, except with different values for the settings because you have a different CPU. We'll keep the ram relaxed at this point, to keep it out of the picture. Once AI Overclock Tuner is set to manual, all the other settings become available. AI Overclock Tuner - Manual CPU Ratio Control - Manual Ratio Setting - 9.5 FSB Frequency - 333 FSB Strap to NB - Auto PCIe Frequency - 100 *** never change this! DRAM Frequency - DDR2-800 (we'll increase this later) DRAM Command Rate - 2T DRAM Timing Control - Manual 5 5 5 15 (we'll try tighter timings later) the rest auto That's it! Boot to memtest and see if the bandwidth numbers changed, up or down? If it tests OK, boot to the OS and check cpuz and memset, did any of the ram timings or speed change? Up or down? Feel free to post your results.

The CPUs these days are quite capable, and it can make a huge difference in the way the machine feels. Granted, the faster the stock speed of the CPU, then less the percentage of performance increase, like a 3.1GHz E8500 overclcoked to 4.1GHz is a 1GHz improvement, but a 2GHz E2180 overclcoked to 3.2GHz is a 1200MHz improvement. It represents like a 30% improvement as opposed to 60% - my math right? Telling it to go faster in the BIOS is exactly what you are doing! It's just not a single setting like it is in the new X58 and core i7. You do need to have a basic understanding of how the system works, in order to make logical and correct descisions of what to change. As basic as it gets, it is just the CPU multiplier times the Frontside Bus speed, equals the final CPU speed. Ram speed is set with another muiltiplier that is actually a divider. So it is independant on the CPU speed, and you can run slow ram with a very fast CPU -- you will just be limited as to how far it can go because eventually your ram can run out of speed on it's lowest divider. As well, Ram has timings, or latency settings that can improve performance when set tighter (lower numbers), or relax it to run faster with set loose (higher numbers). The great thing is that fully auto settings in these fields are (these days) just fine. CPU Multi x FSB speed = CPU speed Ram divider x FSB speed x 2 (for DDR) = RAM speed One confusing thing is that DDR speed is twice the actual ram speed. Even more confusing, is that the FSB speed on Intel socket 775 is "Quad Pumped", meaning that it is multiplied four times to get it's final speed. So a 1333MHz FSB is actually 333MHz times 4. In most BIOSs, you would enter 333 in the FSB field for a 1333MHz FSB, but a good BIOS will tell you what your final speeds are after entering any number in the FSB field. Either way, keep a calculator handy! So, Blair's first CPUZ shot shows us that his CPU is running 3172.8MHz (or 3.17GHz rounded). The Multiplier is 9.5x. The Bus speed is at 334.0MHz (333MHz rounded). His Rated FSB is then 1335.9MHz (1333MHz rounded). Don't worry about the odd numbers being reported, that is normal: 333MHz x 9.5 = 3163.5MHz or 3.16GHz. The multiplier on these CPUs is locked, you cannot adjust them upward. You can adjust them downward though. This may come into play later, but for the most part the highest available CPU multi is best. So the only way we can tell the PC to go faster, is to increase the FSB speed. That is simple enough, once the feature is unlocked by changing that first setting... lol. Lets say we change Blairs FSB to 350MHz, that gives us: 350 x 9.5 = 3325MHz or 3.33GHz rounded. Interesting, because intel has a model that runs at that speed, but gets there by a higher multiplier: 333 x 10 = 3.33GHz The E8400 gets a 9x multi: 333 x 9 = 2997MHZ or 3GHz rounded! See how it works? Questions? OK, so school teacher here: Blair wanted a 3.5GHz overclock, what would his FSB need to be to get that? I can't believe with a machine that big, OC is even considered, let alone needed at this point. I'm barely even half that, and can play this game perfectly.

Yeah, more to come, I have some time today. Don't let it scare you off, it is a little wordy but the steps go fast and are easy to do. It becomes very routine after a few sessions. It helps to have another system to use when it is benching and stressing, or do some housework. Set your boot order to boot to the CD Rom only! Simply disable anything below it in the list. As it is, it is probably asking you to hit any key to boot to CD, and you either are missing the message or your screen hasn't come up as fast as the system boots. Force it to boot to the CD, then it should work. Going back to the HD is easy, change the first (only) boot device to your hard drive then save and exit. I run all my systems like this, it speeds up the boot.

Intro to MemTest I think it is important to know the system at stock speed is stable, and what sort of performance numbers you get, as well as what the AUTO settings actually set things at. CPUZ tells us a lot, but only some basics compared to what is available in your BIOS. That's why you need MemSet for the ram, and some sort of sensor program for sensor data. Using the stockspeed as a baseline will show you how much improvement you get. So the first column of the worksheet, I put in full auto settings. The second column is usually stock speed but manual settings, and perhaps tighter ram timings and faster ram speed. Unless you know these are stable, overclocking can be a nightmare. So the first thing to do for you I think, is to reset your BIOS to defaults, then lean it out by disabling things you don't need or have (this will improve boot times, and you will be rebooting a lot). In the last page of the BIOS, select Load System Defaults or Load Optimal Defaults. Reboot, and return to the BIOS. Turn off things like Floppy, parallel ports, serial ports, IDE, anything you don't use. Go to the boot menu, and make the first boot device the optical drive, disable all the others. Put the Memtest disk in the drive. In the Power menu, Hardware Monitor, disable Qfan. In AI Tweaker, everything to auto, except disable the spread sectrums. In the Advanced menu, CPU Config, check that these are so: Ratio Control: auto C1E: enable CPU TM: enable Vanderpool: enable Execute Disable Bit: enable Max CPU Limit: disable Intel Speedstep: enable Later, we'll have to change a few of them. Reboot, and return to the BIOS. Lastly, in the Tools menu, Asus O.C. Profile, save the settings to Profile 1. This is your fully stock settings. Now, reboot and let it boot to the memtest disk. In the upper left side of the screen there are a bunch of numbers. These indicate calculated bandwidth of the various memory in the system. I record these in the first top boxes of the worksheet, labled BIOS Version, Memory type, etc. The first line of numbers is the CPU speed. The second line is the Level 1 Cache speed. The third is the Level 2 Cache speed. The fourth is the main memory speed. It's the main memory speed you really want to watch. The cache speeds are locked to the cpu, the faster the cpu goes the faster the cache. The RAM speed is dependant on the memory divider you are using, so doesn't have to scale with the cpu speed. Record these on your worksheet, and let it run a few full passes. If there are any errors, they will show uop as red text in the lower part of the screen. Hopefully, you see no red text. Now lets try a few passes of only test 5, a very good test for memory problems. We want to run at least 20 passes for test 5. Hit these keys: C, 1, 3, 5, enter, enter. Now it will continue to run only test 5, until you turn off the computer. You can do that with any test 1 through 0. Have a look at some of the other menu options in there, but there isn't much else I regularly use. If there is no red text after 20 or so passes, reboot and enter the BIOS. Change the boot device to the hard drive, save and exit. Once in the OS, open memset and see what your ram timings are, record these in the lower section of the worksheet. Open cpuz, and record the memory timings in the middle section of the worksheet and the basic CPU stuff in the upper section. Anything you are not sure of, leave blank. It may not be pertinent yet, or available. Once overclocked, things start coming off auto real fast. Run all the benchmarks you want, and record the results as best you can. I usually just take screenshots of them, and keep them in a folder called stockspeed yadda yadda. Sometimes it's good to have cpuz and memset open in the same shot, so you know what the system was set to when the bench was done. I prefer to just keep the screens in different folders named at the speed the system is at. In the next post, we can take things off auto but stay at stock speed. The first thing to know about manual settings is how to set it up at stock speed.

Here are some worksheets I make and use to track and record settings. I three hole punch them and keep them in a binder with any other info I print out for the system. One of these two should be a good starting point for the P5E, similar enough anyway. I'll make a custom one for you later. http://members.shaw.ca/falcon_icehole/AsusP5B-Worksheet.pdf http://members.shaw.ca/falcon_icehole/AsusP5K-Worksheet.pdf This is a rough sheet I use, usually only putting important milestones in the other dedicated one. This one has more room for a progression of settings when trying to settle something down. http://members.shaw.ca/falcon_icehole/memtest-worksheet.pdf This is a generic sheet, works for just about any nForce chipset and many others. http://members.shaw.ca/falcon_icehole/Generic-Worksheet.zip This is a temperature workseet, great for testing airflow configurations or new coolers. http://members.shaw.ca/falcon_icehole/Temperature-Worksheet.pdf

Get these additional apps we'll need (these are a must): Memtest - burn a CD of the ISO here: http://www.memtest86.com/ CPUz 150 is out: http://www.cpuid.com/cpuz.php PC Wizard: http://www.cpuid.com/pcwizard.php MemSet: http://www.xtremesystems.org/forums/showthread.php?t=92190 Orthos: http://www.overclock.net/downloads/138142-orthos.html Prime 95: http://majorgeeks.com/Prime95_d4363.html HyperPi (SuperPi on Steroids - for stability and benching): http://discussions.hardwarecentral.com/showthread.php?t=179504 These are additional (if interested): Sandra (for benching - grab the free version): http://www.sisoftware.net/index.html?dir=&location=downandbuy&langx=en&a= Everest Ultimate (for benching and sensors, sys info - grab the trial version of Ultimate): http://www.lavalys.com/products.php?lang=en&page=11 Speedfan (sensors and fan speed control): http://www.almico.com/sfdownload.php WebTemp (graphing sensor data over time, use with SpeedFan): http://www.webtemp.org/ Additional Benchmarks 3DMarks - hard to find the earlier non online versions, use whatever free version avail. http://www.futuremark.com/ Aquamark - Very quick to run bench with great simple data feedback. http://downloads.guru3d.com/download.php?det=673 Utilities Screenshot Captor: http://www.donationcoder.com/Software/Mouser/screenshotcaptor/index.html Irfanview - Screenshot resize and conversion, batch capable, free, very fast. http://www.irfanview.ca/

The Extreme Tweaker settings will become unlocked when you set the AI Overclock Tuner to MANUAL. I think... it may be another one... lol. I'll check into the BIOS later. It's not a big deal yet, and at least we know what you are on. Sorry for the confusion, I was looking at your last cpuz shot, and the ram is listed as PC2-5300, but also that it's part number is OCZ2RPR11501G, the same as the link I posted only that it isn't a 2gb kit and is only a single stick (xxxx1G as opposed to xxxx2GK as in the link). It also has the EPP profiles for 1150MHz operation. Do you have one or two sticks of this in your system? OCZs part numbers are easy to decipher: OCZ - obvious 2 - DDR type, DDR1 DDR2 DDR3 etc RPR - Reaper HeatPipe Conduit (HPC) 1150 - rated MHz 1G - size of density, 1Gigabyte The link has a K on the end, indicating a 2 stick kit. It seems the SPD data is wierd in some way, either you got a really nice stick of ram rated at a much lower speed, or the SPD data is incorrect for what you have. Both are common things sort of. Is your ram the OCZ Reaper Heatpipe Conduit? I didn't know they made it at PC2-5300. Do you know the exact models you have? You can pull one and read the sticker on it, part number starts with OCZ... That's the rated ram voltage in that last screenshot, what the manufacturer recommends or requires. Not what the system is actually set to. I'll show you where to set the voltage inna bit. Your ram is running at 1066MHz, not 333. As per the third screenshot, the first ram one, look at DRAM Frequency, it says 534.4MHz. Round that to 533, and then double it to account for DDR (double data rate) and you get 1066MHz. Don't ask me why, it's just a fact that a lot of programs refer to ram in it's actual speed and not it's DDR speed.

Nice ram! http://www.ncix.com/products/?sku=23484&vpn=OCZ2RPR11502GK&manufacture=OCZ%20Technology What voltage are you running them at? The sensor data doesn't show vdimm.

The good thing is that fan is super easy to remove and put back on. After doing it a few times, I am sure anyone could do a re&re in less than 30 seconds. As long as the fins are lined up, it snaps on with such dexterity and lack of force that you'd think it wasn't on properly. To take it off, just disconnect the fan tail and pry one side's clip up a bit with a fingernail, off she comes. If there is a lot of dust on it, just suck the big chunks off first with a vacuum. P5E is nice! Reading up here first: http://www.xbitlabs.com/articles/mainboards/display/asus-p5e.html I think you will need to check your BIOS version, possibly need to update that. The X38 chipset while maybe dated these days it was top of the line. The E8500 should get close to or beyond 4GHz on air, any reason you chose 3.5ghz? The reason I ask is twofold: Firstly, I've always been an advocate of setting a goal speed and then working with the system to get to that goal. Bench it and stability test it, set a new goal, and repeat. I like this method as it is fun to find out what it can do when you get there, rather than already knowing. It's a longer process in the end, more for savoring rather than rushing through, and it is a great way to learn out the settings. Sort of the journey not the destination scenario. This is just sort of the way I have personally done it in the past, even though I have seen others doing it this way it is not really a documented method. Secondly, the most popular method of overclocking is to separately test the CPU, MOBO and RAM to find their maximum stable operation speed, then put it all together in a final setting, bench and test it. Routine for skilled overclockers, it is the quickest path to the end. For the unskilled, it can be often (except for the benchmarks) unstable. Anyways, when you come to learn the second method, you will approach the first method differently. Rather than just picking a speed, we calculate things a bit and go for a number that is beneficial in some way. For example, if your cpu stock fsb is 800, then a logical progression would be to see if it runs at 1066MHZ or 1333MHz. Since your cpu is stockspeed 1333MHz, your first logical step is to 1600MHz. With your multiplier at stock maximum (9.5), this puts your cpu speed at 3.8GHz. That should be your first goal, 3.8GHz. At first keeping the ram running slower, and when stable then try speeding the ram up. Ram is the real fun part, a lot more time goes into getting the ram stable but at it's fastest capable speed. There are way more ram settings to be tried than cpu/fsb settings. I've found that some Asus boards have a setting in the BIOS that severly limit the overclock, I will try to dig this up from a P5B I have. There are some other settings in the BIOS that should be made as well. You need to turn off your power savings features, like C1E or CPU TM, and Spread Spectrum. I've got some PDF worksheets for recording your progressive settings, better than a desk full of scrap paper notes. I'll post one that is close to the P5E bios once I get familiar with it. I'll BBL when I read the review and look at the bios.

OMW there now!

For me, the most enjoyable part of assembly is the cable management. I am not as anal about it as others, and often prefer quick access to things rather than making a system look like it doesn't have any cables. In the end, the system looks like it was professionally wired by an electrician specializing in industrial buildings. Temperatures are surely affected, most noticable when rewiring a system that was produced by a mom and pop store, or large vendor like Compaq, HP or Dell. Back in the old days, airflow was not an issue, and most systems had no fans other than the PSU and CPU. The cables were invariably a rats nest that was hit by a tornado! No wonder most people feared upgrading their system or the DIY build, because how could one know what is connected to what, when you cannot even see the CPU below the mess of wires. Things are much better these days, where systems from Dell and the like have much more thought put into the cables. While they still leave stuff accessable, the cases come with fans and airflow is present, and the systems are still usually very very quiet. Cases are made with much thought to cable management, with holes and slots made for routing wire, lots of premounted fans, and unrestricted intake and exhaust venting. There are two schools of thought on a system these days. Both have enthusiast level of following. One is the optimal cooling and performance, the other is the completely silent PC with optimal power savings. Both schools are very interesting, and a joy to look at what people have done. There is also a miriad of inbetween! Great stuff! There is also Old School and Contemporary. Old School would be an older system with ribbon IDE cables etc, Contemporary would be using the current state of technology. You can see how things were when this was just starting, like rounded cables or ribbon cables carefully folded so they do not restrict airflow. I've always believed the saying that for one to learn the future, one only need study the past. I'll try to find some time to sift through my pictures and post some examples of my work. I've also run into some great examples of others work that I can maybe link to.

Interesting indeed! I've not yet heard of such a thing, although I suppose it is possible. Perhaps they have done something related to power that requires PCIe 2.0. Regardless, if it requires a PCIe 2.0 slot in single card mode, then SLI or Crossfire problems would not be an issue. Also, while issues with running two PCIe 2.0 cards in SLI or CF and in PCIe 1.0 slots are known, they are not documented in any product's specs or manuals that I know of. I have only read about the issues from reviews or user experiences.

Can't get there, off limits error.

Yes, positive. All new PCIe 2.0 vid cards are backward compatible with PCIe 1.0 slots. It was very necessary to do this. It's only when you try to run SLI or Crossfire that you may encounter problems. Performance boost for the higher bandwidth PCIe 2.0 slots is negligable, from what I have seen with benchmarks and reviews. My 9800GTX runs fine in my eVGA 680i, which is PCIe 1.0. Are you sure about this Falconise? - AKAIK my NV card won't run in a PCIe 1.0 slot. Admittedly I've not tried it, but then again testing anything other than the recommended solution with 180UKPs-worth of hardware is not my idea of a Good Plan.

The biggest difference between PCIe 1.0 and PCIe 2.0 is the data bus is wider. That hardly matters if the vid card is far from saturating the bus in PCIe 1.0. Another difference is the power that PCIe 2.0 can supply through the slot is higher, meaning there is less need for the separate power connection to the vid card. I believe also, that power savings states are better supported in PCIe 2.0, meaning that the system can actively lower the power consumption of the card, and/or completely shut it down when not needed (thereby using integrated video, when supported). PCIe 2.0 won't come into play performance wise, unless running two cards in SLI or Crossfire. You can run a PCIe 2.0 vid card in a PCIe 1.0 slot, with little or no hit to performance. You can run a PCIe 1.0 vid card in a PCIe 2.0 slot with no difference at all. You can run dual PCIe 2.0 cards in PCIe 2.0 slots, but you will have problems running two PCIe 2.0 cards in PCIe 1.0 slots.

Yes indeed! The vid card's ram is completely separate from the system, and data streams go through the PCIe bus to and from system memory to the vid card memory. Once the data is in the vid card, it is manipulated internally with 3d stuff and the vid card processor. The faster vid card ram is able to do this manipulation better, to and from the vid card processor. The latest high end ATI cards are using DDR5 ram, (dunno what happened to DDR4 it was skipped or something). However, while the DDR5 ram helps the ATI cards, it isn't a significant enough boost to warrant using it in the NV cards yet, which are able to keep up with the ATIs while using DDR3.

Probably happened automatically when they adjusted the vid card price, since the free shipping offer ended on the mobo. They should fix no problemo.

Their service departments is visible from the front desk at the location I pickup from. I've always been impressed at the attention they give to builds. Looking at the part boxes stacked up for the builds, these are not cookie cutter systems, but all quality stuff! Cable management is impressive. One of their employees has been doing DIY videos that are now on a youtube page they have. Unlike other outlets, they do care about what they are shipping people, and are quite knowledgable about anything an enthusiast would care about. Considering they offer an overclocking service where you can order your system pre-overclocked and stress tested with a warranty, I feel they are one of the few Canadian outlets that actually know what they are doing. Here is their Youtube page: http://www.youtube.com/user/NCIXcom This Linus Sebastian is a young strapper, but that's what you want in this business!

Awesome!!

That is sort of what I would do. I don't mind being one step behind the cutting edge, lets you stay in top of the line components at a much lower price, the system is still very powerful, and I can never have enough spare systems. By the time I go into a core i7 system, it'll cost me much less than todays prices. Intel just released a new stepping for the core i7, and it appears a big improvement on power requirements and overclocking. And considering the Rampage II board is on Rev 2, buying into the core i7 later rather than now, only gets you a better product in the end.

Oh, another thing about the Rampage II is that it is an eATX size, which is bigger than the standard ATX (usually a server board is this size). So it will only fit in a larger case that has an eATX spec. Examples are the Cooler Master HAF932: http://www.ncix.com/products/index.php?sku=33029&vpn=RC%2D932%2DKKN1&manufacture=COOLERMASTER&promoid=1001 The Cosmos, and a number of others: http://www.ncix.com/products/index.php?minorcatid=104&subminorcatid=88 .

Glad to see this! I've heard it before that NCIX and DC doesn't mess around when it comes to shipping. For folks that live in an area that may not have a decent computer supplier, this is a godsend. Buying the VIP membership enhances this with free RMA returns and priority queue for orders. The price search is a huge advantage and time saver, as long as you don't get tripped up with false prices like we did above. You just have to follow the links farther and verify prices, but a few clicks later you can be sure you have the best price in Canada. That Rampage II is shweeet! It's the one I would like to get when I get my next upgrade, mostly for the overclocking dials. Otherwise, the eVGA X58 is my next choice because I won't have easy access to those dials when the system is in my case. They have now released a version 2 of the Rampage, so I would safely say that the hardware bugs have been fixed. I mean, bugs that were present in the first board that were not fixable by software or BIOS updates. This always seems to happen with any cutting edge boards, and can sometimes be huge problems. I remember eVGA replacing (for free) a lot of bad 680i boards (iirc). The Rampage II will not run an E8xxx cpu. It is a LGA 1366 socket, not a socket 775. So it only takes the Core i7 CPUs. The Core i7 920 is the cheapest of which, at around $350. Also, it is a DDR3 only board, and takes Triple Channel so you need a kit of THREE sticks DDR3 memory, which is still a little pricey for the good stuff. The good thing is that memory performance is out of this world, and the CPUs finally have a built in memory controller for insanely low latency. Cutting edge stuff, not cheap to get into, but well worth it if you regularly take advantage of 4 cores with video encoding etc. Have a look here for some expert info and benchmarks: http://i4memory.com/f54/ Same guy, with a Core i7 running at 4.7GHz on AIR! http://i4memory.com/f80/i7-920-4769mhz-true120-air-13746/ Probably gobblygook to most folks, but to get a perspective on things look at his SuperPi benchmark in one of the screenshots. He has completed a SuperPi 1M test in 8.6 seconds with the CPU at almost 4.8GHz. That test would take over 30 seconds to complete with an AMD X2-6000 at 3GHz. This is insanely fast!

The more memory you have, the longer it takes. A single full pass of 2GB of fast ram in a fast system may take 30 minutes to an hour. Double that for 4GB of ram. Running a single full pass is just a starting point. I usually run a few full passes, then 20 passes of test 5. Repeat for every change I make in the BIOS. See how it can start consuming time? You had one dimm fail? In what test? It may not be the dimm, it may just be a setting, or perhaps swapping the sticks in the slots may help. First thing to do is pull the good stick and run just the one that errored. Run a full pass, see what happens. If it still errors, you can then tweak the timings settings to see if you can settle it down... it really depends on what test it errored on. If you settle it down, then add the second stick and see if it is still ok. If we can determine that the stick is bad, you are best to not run it at all, it can hose your operating system. Memtest may seem quite simple at first glance, but it really is a very powerful tool when used right, and has a lot of options hidden away. For the most part, it is used in a process of elimination to determine problems or perhaps if changes made in the BIOS are stable and improve performance or not. When overclocking, passing memtest is no garuntee that the system will be stable in the OS. However, if memtest errors you can be damn sure the OS will not load safely. You can leave overclocking alone, or until you get interested in it. Power savings and sleep functions work best with stock settings anyway. However, stock settings don't necessarily mean AUTO. You can run stock speed with manual settings and get a pretty good boost in performance. Tweaking the memory falls under this. For example, you purchased DDR2-1066 memory that is rated to run at 1066MHz. However, the mobo when set to auto will most likely run the ram at 800MHz (which is the industry standard speed for DDR2 ram). To get the ram running at 1066MHz you will have to make some changes in the BIOS, and then verify that things are stable with Memtest, and then in the OS by running some stress programs. If the settings are not stable, you can hose your OS install real easy, and at the very least the computer will crash or bluescreen.

Whoaw... that vid card is on sale this week (starting last night) for $150 (-20 MIR). http://www.ncix.com/products/index.php?sku=36773&promoid=1001 May be able to get this new price?