Computing desk
< October 23	<< Sep \| October \| Nov >>	October 25 >

Welcome to the Wikipedia Computing Reference Desk Archives
The page you are currently viewing is an archive page. While you can leave answers for any questions shown below, please ask new questions on one of the current reference desk pages.

October 24[edit]

Do KVM switches allow for simultaneous display[edit]

Do KVM switches normally allow for simultaneously displaying the video output of two computers? That is, would it just divide the screen in two (or more) regions at the same time? Or will they just allow to jump from one video output to the other, but not display them side to side? --B8-tome (talk) 13:09, 24 October 2017 (UTC)[reply]

My KVM switch swaps between using the keyboard, mouse and screen for one computer and for the other. It cannot display 2 "screens" at once. If it did, where would you send the keyboard and mouse input? --Phil Holmes (talk) 14:11, 24 October 2017 (UTC)[reply]

Split-screen KVM switches exist, but they're not cheap at all. here is one for more than three grand, and here's one for almost two grand.

I think the expense is all in the technology that combines multiple live video streams.

Your best bet would probably be to get a multi-view box like this one or a TV that has built-in picture-in-picture support, and then use a normal KM switch.

(Although, really, the absolute easiest way to do this, is just to put two monitors side-by-side. And use a km switch for your keyboard/mouse.) ApLundell (talk) 16:19, 24 October 2017 (UTC)[reply]

Agreed. I like the two monitor solution, because then you also have a backup if one fails, so can at least view one source at a time. With both monitors working, you can still use the KVM switch to share one keyboard and mouse. For example, this would be good to view videos on one, while surfing the web on the other. StuRat (talk) 16:31, 24 October 2017 (UTC)[reply]

You could connect all computers to the network and then manage them through Remote Desktop Services. Without any extra hardware.Hofhof (talk) 20:51, 24 October 2017 (UTC)[reply]

You can do this much more cheaply if you find a display (rather than a switch) which can display two video signals simultaneously. Andy Dingley (talk) 21:16, 24 October 2017 (UTC)[reply]

Launch date of NVDIA GeForce GT 420[edit]

What is the launch date of NVDIA GeForce GT 420 video card?

NVIDIA page about that model: http://www.nvidia.com/object/product-geforce-gt-420-oem-us.html
See also List of Nvidia graphics processing units

I didn't find it in these pages. Did I miss anything? Thanks in advance. --Daniel Carrero (talk) 15:48, 24 October 2017 (UTC)[reply]

@Daniel Carrero: The second link you provided says it was released on September 3, 2010. RudolfRed (talk) 01:31, 25 October 2017 (UTC)[reply]

Oops, sorry for the trouble, I should have seen that. Thanks. --Daniel Carrero (talk) 01:43, 25 October 2017 (UTC)[reply]

Stupid question: Is the Intel GPU running when an NVIDIA or AMD one is?[edit]

Graphics processing is highly parallel so why not, every little bit helps. Sagittarian Milky Way (talk) 17:53, 24 October 2017 (UTC)[reply]

Not if the overhead of coordination and task distribution is bigger than the contribution of the secondary system. Also, software complexity might be an issue. --Stephan Schulz (talk) 19:59, 24 October 2017 (UTC)[reply]

Performance can be a funny thing. I often use two monitors, one running a CAD program on my NVIDIA card and another running a text editor on my Intel GPU. I tried playing a game on the Intel GPU, and found that the CAD program on the NVIDIA GPU slowed down a tiny bit. It turns out that the extra heat from the Intel graphics stopped the CPU from boosting the clock frequency quite as high as it could with the Intel GPU shut off. --Guy Macon (talk) 02:49, 25 October 2017 (UTC)[reply]

Yes, they just don't seem to be designed to work together. StuRat (talk) 03:18, 25 October 2017 (UTC)[reply]

Someone correct me if I'm wrong, but I don't think there's any infrastructure on the motherboard that would allow those two GPUs to communicate with the kind of bandwidth they'd need to make that work.

SLI requires special bus to bridge the two GPUs. ApLundell (talk) 15:00, 27 October 2017 (UTC)[reply]

They wouldn't necessarily need to communicate with each other. For example, if playing video in a portion of the screen, one graphics card could work to decode/decompress the video, while the other handles the display of everything else on the screen. Of course, there would need to be some interaction on the motherboard for this to work, but not much. StuRat (talk) 16:47, 27 October 2017 (UTC)[reply]

Download schema from URL in WSDL[edit]

I have a web service with a WSDL file containing a schema. When I publish the web service and access the WSDL from a browser by writing the web service URL and appending ?wsdl, I can see the WSDL with the schema embedded into it. I'd like the WSDL instead include a URL where I could download the schema as a separate XSD file instead.

So instead of the WSDL file showing something like this:

<xs:schema>
  <xs:element name="foo" type="tns:foo">
  </xs:element>
</xs:schema>

I'd like to have it show something like this:

<xs:schema url="http://localhost:8080/myservice/myservice.xsd"/>

so I could type http://localhost:8080/myservice/myservice.xsd into a browser and download a separate file containing the <xs:schema> definition above. How is this possible? JIP | Talk 21:03, 24 October 2017 (UTC)[reply]

Update: I am using the Apache CXF framework to publish the web service and am deploying it under Apache Tomcat. I've read online that it should be possible to have the schema file accessible with a URL like http://localhost:8080/myservice?xsd=1 but this gives me the following error:

<soap:Envelope><soap:Body><soap:Fault><faultcode>soap:Server</faultcode><faultstring>Could not find xsd 1</faultstring></soap:Fault></soap:Body></soap:Envelope>

JIP | Talk 22:35, 24 October 2017 (UTC)[reply]