Return to Answer

My guess would be, the KVM is a purely passive device that supports and supports only the DP standard (including dual-mode, a.k.a. DP++), therefore it will work with:

1. connection involve DP and DP only (obviously)
2. a DP++ source (PC) and a TMDS sink (Monitor), i.e. [PC]DP->DP[KVM]DP->HDMI/DVI[Monitor]

(NOTE: actually the second case probably will not require any explicit support in the KVM as the source itself is responsible for the detection; as long as the KVM, or you, doesn't do anything silly in the middle that blocks/hinders such detection, it will work)

The problem probably lies in the HDMI-to-DP adapter/cable. Unlike DP(++-to-HDMI), there's no such thing as "passive" HDMI-to-DP conversion. I cannot say for sure which kind of adapter/cable you got, but what I suspect now is, the one has chip to do TMDS to DP signal conversion.

That's the reason why the "diagram 3" case work, because with an "active" cable, from the KVM/sink's point of view, it is the same case as the first one I listed above.

Now let's back to the case where it doesn't work. For a TMDS sink (the DVI monitor in this case) to work, the source side must be a DP++ port, which can detect it for being such a sink and send TMDS signal instead of DP signal.

So for your desired connection work, you'll need a cable so adaptive that it works like DP++ port, that is, it needs to support TMDS sink and passthrough the TMDS signal it receives when it detects such sink being connected to it.

The thing is I doubt that such adapter chip/solution ever existed on the market, as it's not exactly something sensical to support / invest in except in corner case like yours. (Although it does not really violate any standard and should be theoretically feasible, as far as I'm concerned; just that I don't think any vendor would bother to produce such product.)

Now someone may say, what about a "passive" HDMI-to-DP cable? Sorry, such thing does not exist either (what exists is DP(++)-to-HDMI), as there's no "HDMI++" that allows DP signal to be transmitted over HDMI, and just as I'm telling, it won't help in your not-working case either even if it exists in some parallel world. It would at best save you the active adapter/cable/signal conversion in the diagram 3 case.

So my conclusion is, you'll need either the KVM or the monitor or the graphic card/motherboard replaced.

TL;DR, the DP-to-DVI adapter/cable(s) you have are most likely passive one(s). Get an active one (for the monitor that is connected to the HDMI source / active HDMI-to-DP adapter), it will likely get things work. (Example that has "active" clearly stated on the packaging)

My guess would be, the KVM is a purely passive device that supports and supports only the DP standard (including dual-mode, a.k.a. DP++), therefore it will work with:

1. connection involve DP and DP only (obviously)
2. a DP++ source (PC) and a TMDS sink (Monitor), i.e. [PC]DP->DP[KVM]DP->HDMI/DVI[Monitor]

(NOTE: actually the second case probably will not require any explicit support in the KVM as the source itself is responsible for the detection; as long as the KVM, or you, doesn't do anything silly in the middle that blocks/hinders such detection, it will work)

The problem probably lies in the HDMI-to-DP adapter/cable. Unlike DP(++-to-HDMI), there's no such thing as "passive" HDMI-to-DP conversion. I cannot say for sure which kind of adapter/cable you got, but what I suspect now is, the one has chip to do TMDS to DP signal conversion.

That's the reason why the "diagram 3" case work, because with an "active" cable, from the KVM/sink's point of view, it is the same case as the first one I listed above.

Now let's back to the case where it doesn't work. For a TMDS sink (the DVI monitor in this case) to work, the source side must be a DP++ port, which can detect it for being such a sink and send TMDS signal instead of DP signal.

So for your desired connection to work, you'll need a cable so adaptive that it works like DP++ port, that is, it needs to support TMDS sink and passthrough the TMDS signal it receives when it detects such sink being connected to it.

The thing is I doubt that such adapter chip/solution ever existed on the market, as it's not exactly something sensical to support / invest in except in corner case like yours. (Although it does not really violate any standard and should be theoretically feasible, as far as I'm concerned; just that I don't think any vendor would bother to produce such product.)

Now someone may say, what about a "passive" HDMI-to-DP cable? Sorry, such thing does not exist either (what exists is DP(++)-to-HDMI), as there's no "HDMI++" that allows DP signal to be transmitted over HDMI, and just as I'm telling, it won't help in your not-working case either even if it exists in some parallel world. It would at best save you the active adapter/cable/signal conversion in the diagram 3 case.

So my conclusion is, you'll need either the KVM or the monitor or the graphic card/motherboard replaced.

TL;DR, the DP-to-DVI adapter/cable(s) you have are most likely passive one(s). Get an active one (for the monitor that is connected to the HDMI source / active HDMI-to-DP adapter), it will likely get things work. (Example that has "active" clearly stated on the packaging) (Another one)

My guess would be, the KVM is a purely passive device that supports and supports only the DP standard (including dual-mode, a.k.a. DP++), therefore it will work with:

1. connection involve DP and DP only (obviously)
2. a DP++ source (PC) and a TMDS sink (Monitor), i.e. [PC]DP->DP[KVM]DP->HDMI/DVI[Monitor]

(NOTE: actually the second case probably will not require any explicit support in the KVM as the source itself is responsible for the detection; as long as the KVM, or you, doesn't do anything silly in the middle that blocks/hinders such detection, it will work)

The problem probably lies in the HDMI-to-DP adapter/cable. Unlike DP(++-to-HDMI), there's no such thing as "passive" HDMI-to-DP conversion. I cannot say for sure which kind of adapter/cable you got, but what I suspect now is, the one has chip to do TMDS to DP signal conversion.

That's the reason why the "diagram 3" case work, because with an "active" cable, from the KVM/sink's point of view, it is the same case as the first one I listed above.

Now let's back to the case where it doesn't work. For a TMDS sink (the DVI monitor in this case) to work, the source side must be a DP++ port, which can detect it for being such a sink and send TMDS signal instead of DP signal.

So for your desired connection work, you'll need a cable so adaptive that it works like DP++ port, that is, it needs to support TMDS sink and passthrough the TMDS signal it receives when it detects such sink being connected to it.

The thing is I doubt that such adapter chip/solution ever existed on the market, as it's not exactly something sensical to support / invest in except in corner case like yours. (Although it does not really violate any standard and should be theoretically feasible, as far as I'm concerned; just that I don't think any vendor would bother to produce such product.)

Now someone may say, what about a "passive" HDMI-to-DP cable? Sorry, such thing does not exist either (what exists is DP(++)-to-HDMI), as there's no "HDMI++" that allows DP signal to be transmitted over HDMI, and just as I'm telling, it won't help in your not-working case either even if it exists in some parallel world. It would at best save you the active adapter/cable/signal conversion in the diagram 3 case.

So my conclusion is, you'll need either the KVM or the monitor or the graphic card/motherboard replaced.

TL;DR, the DP-to-DVI adapter/cable(s) you have are most likely passive one(s). Get an active one (for the monitor that is connected to the HDMI source / active HDMI-to-DP adapter), it will likely get things work.

My guess would be, the KVM is a purely passive device that supports and supports only the DP standard (including dual-mode, a.k.a. DP++), therefore it will work with:

1. connection involve DP and DP only (obviously)
2. a DP++ source (PC) and a TMDS sink (Monitor), i.e. [PC]DP->DP[KVM]DP->HDMI/DVI[Monitor]

(NOTE: actually the second case probably will not require any explicit support in the KVM as the source itself is responsible for the detection; as long as the KVM, or you, doesn't do anything silly in the middle that blocks/hinders such detection, it will work)

The problem probably lies in the HDMI-to-DP adapter/cable. Unlike DP(++-to-HDMI), there's no such thing as "passive" HDMI-to-DP conversion. I cannot say for sure which kind of adapter/cable you got, but what I suspect now is, the one has chip to do TMDS to DP signal conversion.

That's the reason why the "diagram 3" case work, because with an "active" cable, from the KVM/sink's point of view, it is the same case as the first one I listed above.

Now let's back to the case where it doesn't work. For a TMDS sink (the DVI monitor in this case) to work, the source side must be a DP++ port, which can detect it for being such a sink and send TMDS signal instead of DP signal.

So for your desired connection work, you'll need a cable so adaptive that it works like DP++ port, that is, it needs to support TMDS sink and passthrough the TMDS signal it receives when it detects such sink being connected to it.

The thing is I doubt that such adapter chip/solution ever existed on the market, as it's not exactly something sensical to support / invest in except in corner case like yours. (Although it does not really violate any standard and should be theoretically feasible, as far as I'm concerned; just that I don't think any vendor would bother to produce such product.)

Now someone may say, what about a "passive" HDMI-to-DP cable? Sorry, such thing does not exist either (what exists is DP(++)-to-HDMI), as there's no "HDMI++" that allows DP signal to be transmitted over HDMI, and just as I'm telling, it won't help in your not-working case either even if it exists in some parallel world. It would at best save you the active adapter/cable/signal conversion in the diagram 3 case.

So my conclusion is, you'll need either the KVM or the monitor or the graphic card/motherboard replaced.

TL;DR, the DP-to-DVI adapter/cable(s) you have are most likely passive one(s). Get an active one (for the monitor that is connected to the HDMI source / active HDMI-to-DP adapter), it will likely get things work. (Example that has "active" clearly stated on the packaging)

My guess would be, the KVM is a purely passive device that supports and supports only the DP standard (including dual-mode, a.k.a. DP++), therefore it will work with:

1. connection involve DP and DP only (obviously)
2. a DP++ source (PC) and a TMDS sink (Monitor), i.e. [PC]DP->DP[KVM]DP->HDMI/DVI[Monitor]

(NOTE: actually the second case probably will not require any explicit support in the KVM as the source itself is responsible for the detection; as long as the KVM, or you, doesn't do anything silly in the middle that blocks/hinders such detection, it will work)

The problem probably lies in the HDMI-to-DP adapter/cable. Unlike DP(++-to-HDMI), there's no such thing as "passive" HDMI-to-DP conversion. I cannot say for sure which kind of adapter/cable you got, but what I suspect now is, the one has chip to do TMDS to DP signal conversion.

That's the reason why the "diagram 3" case work, because with an "active" cable, from the KVM/sink's point of view, it is the same case as the first one I listed above.

Now let's back to the case where it doesn't work. For a TMDS sink (the DVI monitor in this case) to work, the source side must be a DP++ port, which can detect it for being such a sink and send TMDS signal instead of DP signal.

So for your desired connection work, you'll need a cable so adaptive that it works like DP++ port, that is, it needs to support TMDS sink and passthrough the TMDS signal it receives when it detects such sink being connected to it.

The thing is I doubt that such adapter chip/solution ever existed on the market, as it's not exactly something sensical to support / invest in except in corner case like yours. (Although it does not really violate any standard and should be theoretically feasible, as far as I'm concerned; just that I don't think any vendor would bother to produce such product.)

Now someone may say, what about a "passive" HDMI-to-DP cable? Sorry, such thing does not exist either (what exists is DP(++)-to-HDMI), as there's no "HDMI++" that allows DP signal to be transmitted over HDMI, and just as I'm telling, it won't help in your not-working case either even if it exists in some parallel world. It would at best save you the active adapter/cable/signal conversion in the diagram 3 case.

So my conclusion is, you'll need either the KVM or the monitor or the graphic card/motherboard replaced.

My guess would be, the KVM is a purely passive device that supports and supports only the DP standard (including dual-mode, a.k.a. DP++), therefore it will work with:

1. connection involve DP and DP only (obviously)
2. a DP++ source (PC) and a TMDS sink (Monitor), i.e. [PC]DP->DP[KVM]DP->HDMI/DVI[Monitor]

(NOTE: actually the second case probably will not require any explicit support in the KVM as the source itself is responsible for the detection; as long as the KVM, or you, doesn't do anything silly in the middle that blocks/hinders such detection, it will work)

The problem probably lies in the HDMI-to-DP adapter/cable. Unlike DP(++-to-HDMI), there's no such thing as "passive" HDMI-to-DP conversion. I cannot say for sure which kind of adapter/cable you got, but what I suspect now is, the one has chip to do TMDS to DP signal conversion.

That's the reason why the "diagram 3" case work, because with an "active" cable, from the KVM/sink's point of view, it is the same case as the first one I listed above.

Now let's back to the case where it doesn't work. For a TMDS sink (the DVI monitor in this case) to work, the source side must be a DP++ port, which can detect it for being such a sink and send TMDS signal instead of DP signal.

So for your desired connection work, you'll need a cable so adaptive that it works like DP++ port, that is, it needs to support TMDS sink and passthrough the TMDS signal it receives when it detects such sink being connected to it.

The thing is I doubt that such adapter chip/solution ever existed on the market, as it's not exactly something sensical to support / invest in except in corner case like yours. (Although it does not really violate any standard and should be theoretically feasible, as far as I'm concerned; just that I don't think any vendor would bother to produce such product.)

Now someone may say, what about a "passive" HDMI-to-DP cable? Sorry, such thing does not exist either (what exists is DP(++)-to-HDMI), as there's no "HDMI++" that allows DP signal to be transmitted over HDMI, and just as I'm telling, it won't help in your not-working case either even if it exists in some parallel world. It would at best save you the active adapter/cable/signal conversion in the diagram 3 case.

So my conclusion is, you'll need either the KVM or the monitor or the graphic card/motherboard replaced.

TL;DR, the DP-to-DVI adapter/cable(s) you have are most likely passive one(s). Get an active one (for the monitor that is connected to the HDMI source / active HDMI-to-DP adapter), it will likely get things work.