Question about multithreading in hardware mode

[paraemuladores]

Hi there. I recently tried to play Justice League Heroes in DirectX11 Hardware mode, but it works a little slow (I have a Core2Quad Q6600 2.4GHz, 3,5 GB RAM, Nvidia N560GTX, Mother Intel DP35DP), unless I use a big amount of Speedhacks, which causes the game to have bad graphics. I remember that I was told here in the forum that I need a better machine to run the game, so, here's the big question. Is there any way to make the emulator use all the four cores, like when in Software Mode you add additional threads? That would be amazing. Not the Speedhack MTVU, but adding all the cores into the emulator, but in hardware mode. Is it possible?

[Nobbs66]

Due to the nature of multithreading and emulation, adding support for more cores is both difficult and possible no faster due to the extra overhead of keeping sync between all cores. So, in short even if it's theoretically possible, you may not see any performance gains with it.

[paraemuladores]

Yea, I was wondering if in order to maintain the sync between the cores, that would happen. I just thought I was wrong because in Software mode you do can add more threads, but I imagine this is possible because absolutely all the emulation is done in the CPU. Thanks for the quick answer.

[gregory]

On the GS side, the game does a lots of CPU-GPU transfer. A faster PCIe link would/could help. Let's hope PCIE Gen 4 is release soon enough.

[Serial Hacker]

On the GS side, the game does a lots of CPU-GPU transfer. A faster PCIe link would/could help. Let's hope PCIE Gen 4 is release soon enough.

so the PCI-E link is a more of limitation now than the card's Bandwidth ?

[gregory]

What is the card bw for you? You mean the link between gddr and the shader? Normally it is much much much faster than the pcie link

[refraction]

so the PCI-E link is a more of limitation now than the card's Bandwidth ?

Hmm not sure, the theoretical bandwidth of a PCIe 3 graphics card is 32GB/sec, which is obviously faster than system ram. That is theoretical however, much like dual channel memory is not a 50% increase in speed.

as a general rule, PCIe 3.0 isn't any faster than PCIe 2.0 which is a theoretical maximum of 16GB/sec, but again that could be a system ram limitation.

[Serial Hacker]

What is the card bw for you? You mean the link between gddr and the shader? Normally it is much much much faster than the pcie link

are you telling me a Faster PCI-E link with no extra bandwidth/clock/performance enhancement from the card side alone would increase performance?

[refraction]

are you telling me a Faster PCI-E link with no extra bandwidth/clock/performance enhancement from the card side alone would increase performance?

The graphics card is not the limitation, GDDR5 can do over 200GB/sec which far outstrips the PCIe link speed and system RAM speeds.

[Serial Hacker]

I Have a GTX 750 Ti and it's showing 86.4 GBps of Bandwidth and i'm sure that these limitations are not related to either my PCI-E link or my Memory type