DR ZĄBER

Console-Sized Gaming PC Chassis

[DR ZĄBER @ZaberSentry]

Another thing just to explain the concept what can be done with standardised graphics card sizes:
https://t.co/Tk3KrBDaEB

[DR ZĄBER @ZaberSentry]

We're starting to show what is next after Sentry: https://t.co/Sr7nXcw8yk
Come and have an opinion :) https://t.co/udwXlwoRsp

[DR ZĄBER @ZaberSentry]

@kfa2gaming Are you perhaps standardising the way your PCI bracket is mounted across different models including future ones? This could go a long way towards interesting cases like the XiKii industry cases, but not locked onto a single model.

[DR ZĄBER @ZaberSentry]

@gcm0rais @SapphireTech If and when there is availability of cards that are not stupidly big and overpriced and power cables for them don't try to burn down your house if you try to squeeze them into the SFF case...

[DR ZĄBER @ZaberSentry]

@aschilling Isn't this actually caused by tilting the card in the opposite way than what ASUS video is showing? And in an open test bench? It seems like a forced removal while grabbing by the far end of the card from the pci bracket which with the tilt can affect that exact pcb corner.

[Hardware Canucks @hardwarecanucks]

I 3D Printed a TINY bracket that made my ITX build SO MUCH better. ❤️

➡️ https://t.co/rOtMaAddEg

[𝐷𝑟. 𝐼𝑎𝑛 𝐶𝑢𝑡𝑟𝑒𝑠𝑠 @IanCutress]

DDR5 is coming, and you've probably heard that it has ECC. That's wrong! It has 'on-die ECC', which is different. Here's the @techtechpotato 💻💻🥔 explainer, and how even manufacturers are screwing it up. Even Linus can't disagree here!

https://t.co/gpZZoy3Gf5

[𝐷𝑟. 𝐼𝑎𝑛 𝐶𝑢𝑡𝑟𝑒𝑠𝑠 @IanCutress]

Processors are 'square', but wafers are round. Lots of wastage, right? Ever wondered why? 💻💻🥔 with the details.

https://t.co/pgPdCdOr8U

[DR ZĄBER @ZaberSentry]

@AsetekDennis The thing is that some boards seemed to default their settings while flashing, so you'll have CPU fan ramped up, but if the pump requires full speed setting and the default is some kind of auto/low or based on chipset temp, then the pump is getting starved. Flashing bios is scary

[Linus Tech Tips @LinusTech]

NEW VIDEO: Maybe upgrading is pointless…..

https://t.co/j04T3K6Hro

#Sponsored by @msiUSA

[Hardware Unboxed @HardwareUnboxed]

Is Zen 3 Really Worth it For Gaming? 5600X vs. 3600 vs. 10400F: https://t.co/2rrVqEnAXT

[optimum @OptimumTechYT]

Ryzen 5000 Undervolting via Precision Boost Overdrive 2 is definitely worth checking out.
Free performance.

https://t.co/EB3nSk5y2J

[DR ZĄBER @ZaberSentry]

@TechCakeMix @AgentM0NARK @linusgsebastian Reference for the joke:
https://t.co/bR1n8uVkK0

[DR ZĄBER @ZaberSentry]

@satansbraten11 Switch to 3.0 barely makes a performance difference and then even switching to 2.0 isn't that much of a drop with cutting bandwidth in half at each drop, suggests the performance drops may come from different 3.0 improvement than just the bandwidth, like for example response time

[DR ZĄBER @ZaberSentry]

Watch this video from start to end.
Note the info about riser vendors going back on stating 4.0 support and recommending to roll back to pci-e 3.0 in bios.

[Hardware Canucks @hardwarecanucks]

There's a PROBLEM with riser cables & vertical GPU brackets! Here's what we found:

➡️https://t.co/2ByIPshHLj

[DR ZĄBER @ZaberSentry]

@satansbraten11 Switch to 3.0 shouldn't really make a problem bandwidth-wise as games are still made in a way that there should be quite a lot of bandwidth left as pcie scaling tests show: https://t.co/Hu1yBjKMR1
Question is whether its a new mechanism introduced in 4.0 or a business decision.

[DR ZĄBER @ZaberSentry]

@hardwarecanucks It seems as when you have multiple x16/x8 slots on the board, this is obvious to have separate switches per slot, but in ITX boards it seems as if vendors try to over-simplify this.

[DR ZĄBER @ZaberSentry]

@hardwarecanucks Would need to dig out the conversation in our thread on https://t.co/TuBlY8HkZC, but at some point according to our clients even high end itx boards had a single switch for everything and we placed a bet that the asrock board would have it separate

[DR ZĄBER @ZaberSentry]

@hardwarecanucks This means there might be a lot of headroom in 3.0 to support direct access memory/accessing SSD from GPU.
Btw if we have this headroom, why not have the output signal routed back to the board displayport/hdmi connector or add-in cards and have tesla-like vent on the pci bracket?

[DR ZĄBER @ZaberSentry]

@hardwarecanucks Also take a look at this: https://t.co/Hu1yBjKMR1
The fact that we still don't have too big of a drop in performance even with 2.0 indicates the slight performance increase comes from other improvements than just bandwidth increase.