Forgive me I am on my phone and it is hard to see. It looks like you have on board video and not a video card the inverted port you see next to it is a serial port. If you want you will have to buy a video card and install it for a second monitor. You would run both monitors of the video card.
Thanks yall- yeah, digging around inside of the tower without it catching fire is not something I think I'm capable of doing. The USB adapter intrigues me, as I'm not looking to do anything fancy with the dual-setup (word processing/data entry/web browsing only- one screen for my local desktop, and the other for my office's remote desktop client) In doing a little bit of reading on this less-expensive Prime alternative, my only concern would be that this adapter is apparently a bit of a resource-hog, and my system would probably meet the bare-minimum requirements. Here is a screenshot of my current specs- do yall see any reason not to drop $40 on this and call it a day?
If you have an available slot inside the computer, just get a second, cheap video card for it. Your computer will then let you use both outputs, and you'll get reasonable performance (way better than USB, which might even be USB1 on that computer, which will be absolute shit for any kind of video resolution or performance). Cheap cards are under $50. $0.02
Blue Dog. Do this. Do not be afraid to open the box. It's very easy. Watch this to see how to open the case: Another option: https://www.discountelectronics.com/product?product_id=3271&product_model=28 https://www.discountelectronics.com/product?product_id=2883
Thanks yall- I'm going to give it a try. For the splitter- my two older monitors can both be connected via VGA and DVI. I've personally only used VGA up until now, if only because I've never had a system that had the DVI outlet before. I should get the DVI-to-DVI splitter, correct? Or just stick with DVI-to-VGA? EDIT- Eff it, nevermind. I only have one DVI cable anyway, so VGA it is. Thanks again for yall's help.
FYI, DVI is better than VGA. DVI gives you more of them there "resolutions" that you can use on the TV. So if you can, don't use VGA.
Rush-O-Matic's suggestion of the ATI HD 2400 XT card with the DVI-to-VGA splitter. I stuck with VGA only because I already had the cables. This was just a test run to see if I could actually do it, though. If I get the approval, I'll probably end up purchasing some new monitors and configuring all of the workstations in my office (in which case I'll splurge for some DVI or HDMI cables).
At my work, we have some simulators that need to be put online somehow for people to use. These simulators vary in size from 5GB to 100GB and are extremely RAM intensive when used on our work laptops, so I'm thinking of doing some sort of 10TB computer people can connect to. Can this be done and any ideas what I can do to make this work? Would something like this be good to use? http://www.newegg.com/Product/Product.aspx?Item=N82E16822165560 Budget is pretty much not a concern, so if I can get something and massively over-build it performance wise, that would be fantastic. My boss told me to figure out what we'd need and make a wish list.
I'm confused. That's just basically a NAS box, isn't it? Are you looking at using the box as a remote server for people to run stuff on, or just a remote drive? What do you mean by "10TB"... disk space? Memory? If it's a simulator, are you sure it's not going to be rather CPU intensive as well? How many people are going to be using it at once? Are you expecting people to still run the simulator process locally using remote drive? How exactly is the simulator interacting with the drive images it's using? If it just reads everything in, processes the fuck out of it, then spits it out to the drive at the end, then that might work. If there's a ton of IO during the simulator, the network latency will kill your performance. Again, I'm quite unclear on what you're trying to accomplish here, and where exactly your problem is.
I am going to need a ton of hard drive space to store the simulators. I don't know how CPU intensive they are, but I know that if we run them on our laptops with only 4 GB of RAM, it freezes our laptops. Basically, I want to keep them on one shared drive or server and allow people to use them through their personal computers as the interface. Almost like playing an online video game where you do your input and actions on your computer, but everything is processed and run on this isolated hardware. Does that make sense?
Exactly this. I'm going to guess that your laptops freezing is not a result of your laptop storage, it'll be either CPU or Memory... and if you run out of memory, it'll then swap locally. I bet it's that local swap that killed your laptops, because when that happens, it's like an order of magnitude reduction in performance, even with SSDs. At that point, you need to run the simulation off of your laptops altogether. Depending on the data you're throwing around, the cloud might not be a feasible option (it takes a fairly significant amount of time to upload/download TBs of data). That sounds like you're now going to build a local machine able to store lots of data as well as run (potentially) more than 1 simulation/user at a time. As Clutch said, you'll need a server-type OS that supports multiple users and remote connectivity, a seriously powerful/mutli-core server-class CPU (waaaay more powerful than the box you linked to), and some serious RAM to handle multiple users at once. Odds are you're looking at more of a local (non-cloud) $5k or so server (on the cheap side), but probably more around $8-10k, not the $1k box you linked to.
I figure the max users that will be on the various simulators at any one time is 5. Currently, we're doing 1 TB external drives to hold the simulators and doing 16GB RAM upgrade kits per person... That's about $300 per person, and there's a potential for 200 users total... I've already explained that this will cost between $7 and $10k, but it beats doing individual hardware upgrades on laptops for $60k. My boss agrees it's smarter to do a $10k machine than spend $60k. That being said, I want to build or buy a killer local machine. I'm a big believer in over-building stuff up front to allow for expansion, etc. The $1k box I linked to was just to see if I was on the right track. If you guys were handling this project and had a $10k budget, what would you go with? As far as access goes, I want to park it on our internal network so that all the users can connect to it remotely. Our security protocols here are on the far side of ridiculous, so I need to have it connected more like a shared drive or something. Any ideas on that? Oh, if it makes any difference, most of the simulators run on VMware.
Personally, I'd go to a local box-builder/pc-shop and tell them what you're looking to do and have them build it for you. Get a big multi-core server class CPU (or 2), get them to load the OS on there for you (I'll make the assumption that it's a Windows OS), and they can work with you to get the best bang for your buck. You should also figure out if you're doing a bunch of high disk activity stuff, not just size of storage, because then you may want to look into getting some high-performance SSD storage, which will drastically speed things up.
Run a simulation on your current machine and take a look at the built-in performance monitor and watch the CPU and disk IO (input/output) and RAM to see what the simulation is doing. Shit that is working hard or spiking will be your choke point. If only your disk IO is pegged and everything else is less than 100%, then you need faster disks (as in SSD). If the disk only is read/write at the beginning and end, and it's huge CPU usage for the vast majority of the process, then you need faster/more CPU's (especially if more than one person will be using the box at a time). If you find all of your RAM gets used right away and the your disk IO starts going nuts, that means you've run out of memory and your disk is now being used as overflow memory (known as "disk swapping"), so you need more RAM. (This is all very simplistic and is for demonstration purposes only). You can also spend a few bucks to optimize your process and simulation, potentially... that might get you bigger returns than a straight up hardware purchase. For instance, if you have big/cheap/slow storage that is like the parking lot for your simulations, and then you have a small but stupidly high performaning SSD that is used just for the running simulation, and then the results are parked back into the old/slow shit, that is a way to get a better end result for cheaper. You can't just throw a bigger box at it and expect it to run better... that doesn't always work. Investigate what is going on, verify where the pain areas are, then come up with a solution that addresses those areas specifically, otherwise it's just a crap shoot. Nothing will suck more than your boss writing a $10k cheque only to have it not fix the problem because you were just guessing and making assumptions.
Yeah, I was thinking of doing all SSD storage anyway, just for the sake of performance. Luckily, I'm not paying for it. With the way they pay for things here, it's easier for me to order it online or build it here instead of a local pc shop. What do you think about something like this? http://www.newegg.com/Product/Product.aspx?Item=9SIA5DJ3C01106
We were all running the simulators on our laptops (500GB hard drives, 4GB of RAM, 2.33mhz processors) and they would freeze. Once we started upgrading RAM (16GB upgrade kits), they all run fine. My current machine is running 32GB of RAM and runs everything like butter.
Yep, that's a server. I just don't know if it will do what you need it to, and that's the price without any drive space or real memory installed. High volume SSD's are expensive. Also, you're going to need a server-capable operating system that will allow multi-user remote logins, which will also affect your price. Again, you're talking some real money here, and if it's your first time digging into this shit, I'd definitely find someone local to help you through the process, who you can better explain your problem to. Do not trust or listen to specific advice on hardware from this forum, or it will probably end up biting you in the ass, because we really don't know your real problem or the specifics of what you're trying to fix... and you're not providing anywhere near the level of detail required for us to provide you with anything other than a general, "it may or may not work" piece of advice. $0.02