themoose5

Fair point, I was mostly just spit balling ideas that would include support for a larger gauge wire all the way though. Honestly if they're increasing the gauge on the wire and connectors to handle 600 watts to the GPU alone the likelihood someone would run a non-modular power supply with that I would say is slim to none. In fact with the current power supply offerings you wouldn't even be able to run a non-modular PSU that could power a system with that card in it. Using the same rule of thumb for power supply capacity as today, a RTX 2080 is recommended to have a 600 watt PSU. The 2080 is rated at 225 watts, roughly 2.6x PSU capacity over the GPU. Translating that to the power draw rating for the new cables. That would be a GPU that's rated for 600 watts and a PSU capacity of 1500 watts. I don't think there is a 1500 watt non-modular power supply out there. Meaning that using an adapter for a card that doesn't need that much power doesn't matter because the card won't be hitting the higher capacities provided by the thicker gauge wires and connectors. If you are hitting that threshold, it won't be on a non-modular PSU and thus could have an included full length cable that takes the spot of the dual 6 pin.

This assumes that this will take the existing dual six pin and just have a short adapter into the card. They could go the route of including a full length cable that would take the spot of the dual 6pin PCIe power connector in the power supply itself. Either way my point is probably in preparation for cards that COULD pull that much power in the future not necessarily for Ampere cards. They are just introducing it now to warm the market up to the new connector before it becomes a requirement for new cards in the future.

I think the quote below is really the motivating driver behind the change. It seems like nvidia is anticipating a much higher power draw/power quality in the future which is why they're introducing this new connector.

There is way too much inertia behind x86 based platforms for a switch like this to happen. There would have to be some MAJOR benefit for everyone involved to really push the market completely away from x86. C++ can't even make ABI breaking changes in part because there are large companies out there that don't have the ability to re-compile important pieces of their software. If something like a language standard can't move forward because of issues like this the death of an entire ISA is going to see even more resistance.

I'm very curious to see how this ends up playing out for MSI. My money's on this is a rouge-ish PR person. I say rouge-ish because almost certainly this person didn't act on their own but as a result of internal pressure or mandates from MSI's higher ups. The internal mandate probably didn't directly say "Pay off reviewers for positive reviews" but I'm sure there is a policy that would reward this kind of behavior by the PR person. Either way extremely suspect when this kind of behavior starts coming out of any company.

I have been thinking a lot about this over the last year or so. I haven't put much legwork into the research but I am very interested by this idea. Specifically seeing if we can increase the distance between keyframes using neural networks to infer the frames between two keyframes.

I haven't followed it closely, what is that status of the Source 2 engine? Would it be feasible that they drop Source 2 and use a 3rd party engine?

Caught the WAN show late but wanted to jump over here to add my 2c. I think that Linus brings up some really good points and some very sad points for myself being a car enthusiast. One thing that I think he's over looking a little bit is this already exists to some extent with options/trim levels that are available with vehicles already. As it is currently a manufacturer will have a chassis or in some cases a shared platform that multiple vehicles are built on. These vehicles will have options available that can be added to the base model if desired. These options add $ to the final cost of the car and are paid for overtime via a lease or loan. Conceptually this isn't that far off from doing what BMW has proposed with just having what is essentially one vehicle with all options included and those options are just gated in a different way. Rather than paying for them to be included in the car, you pay for it over time. I think the only way this ends up working for anyone involved is if the base cost of the car with no options enabled comes down. Now as a consumer I could buy a BMW that maybe I wouldn't otherwise be able to afford but not enable any of the options that come with said vehicle. If I decide to later I can enable them or if not I can keep them disabled. It doesn't take away from what the experience of owning the base model car would be if we stick with the way things are currently done and can open BMW up to new customers. Not that everyone would want to follow this but there are plenty of aftermarket replacements out there that can make this a complete moot point.

Came to this thread from the WAN show topic and wanted to add my 2c. First a little background, I work professionally as an AI engineer. While I don't do direct model development I am very versed in the processes around model development. Also I don't work for google or yt so I don't have specific inside knowledge of their processes or models. On the surface I think Linus' hypothesis he presented on the WAN show could be very close to the truth of how this word ended up on the automatic filter. If in fact google/yt is using a model or collection of models for automated comment filtering it's architecture is probably going to be some thing pretty common.The key here really isn't the architecture of the model but what training data was used to create that model. Specifically where google/yt got that training data from for non-English content. If their training data contained a significant amount of examples containing the Chinese characters in question that lead to a negative sentiment following their use it is easy to see how the model would learn to flag these characters as offensive. This seems to be backed up by the fact that only the use of these specific Chinese characters result in a comment being removed but the use of the romanized version does not receive the same treatment. If this phrase was added by someone manually I think it would be a safe assumption that the person adding them would add in all the variations of the phrase in Chinese characters and their romanized counterparts and not just the Chinese character version. On the other hand if this was something learned by an AI model using data pulled from outside sources all it would take is having enough examples in the training data for the model to learn the Chinese characters of this specific phrase. That learning would not necessarily transfer to learning the romanized version as well, resulting in what seems to be the current behavior. Training data review and analysis is something that is still kind of a sore spot in the field of AI. Training data quality has a large impact on the final performance of a model and there are numerous examples of how AI models can be biased by their training data. Most likely what happened in this case was the non-English language training data was not reviewed well enough to catch these issues before deploying.

Valve is a very unorthodoxly run company so it's hard to say what their strategy is but I'm sure they're not just ignoring what's going on with the Epic store. My bet would be that they're waiting for things to settle down a bit before making big moves. I don't think that the Epic store will be able to keep up it's current pace of offers to developers and deals, including guaranteed sales, for very long. If valve reacts too strongly too early they might end up putting themselves in a bad position in the long run. Better to let things play out a bit and then have a response rather than rushing out of the gate. This is only compounded by the amount of hate that both the Epic store itself is getting as well as developers that announce exclusive deals are getting. It's not like consumers are flocking to the platform. I think it still very much remains to be seen where the Epic store will sit in the PC gaming eco-system.

Anyone had problems using Xbox One controllers via USB after the update? I'm not sure if it's just coincidence but I started getting massive frame shudder issues in a couple of different games that went away when I unplugged the controller after the most recent update a couple of days ago.

Unfortunately the legal system in the US doesn't restrict people who have no right filing a lawsuit from filing said lawsuit. The courts can throw out a claim but that takes some litigation and doesn't stop the suit from being brought in the first place. It's kind of stupid IMO but realistically you can sue pretty much anyone for anything. It might get thrown out but that's a process for the courts and lawyers, there isn't any barrier to entry in that respect besides money. But I agree if their contract had a clause that made it void if Yahoo got sold then this won't take much discussion.

My guess is this is on Verizon for not completing their due diligence. As long as Yahoo didn't actively hide the contract terms from Verizon it's the purchasing company's responsibility to be aware of all the contract terms that the company they're buying are bound by. If Verizon missed this in the Mozilla contract then they have no one to blame but themselves. Though that doesn't stop them from trying via a lawsuit...

That has to be one wicked exponential curve to jump from 16GB to 16TB

I love how advertising corporations are trying to sell the impact to their business as something that is anti-consumer and will hurt end users. Most end users will probably be happy about this not hurt by it. Honestly I don't think anyone in history has ever been hurt because they didn't see a product advertisement.