UltraMega

What are your thoughts on setting up some kind of drop box users can dump video files onto?

https://www.techspot.com/news/101486-microsoft-ai-found-new-material-replace-lithium-li.html Sign of things to come?

You're right the current channel is fine. I was only aware of the podcast channel. Just saw that there is a second one. Looking at a few of the videos on that second channel that have more views, there are no links back to ehw. There are links to review articles but not to the home page. Links to the EHW homepage should be at the top of every video description. Looking at that channel, you can see there is also a clear relationship between thumbnail quality and views. Some of the more recent videos have pretty weak thumbnails and hardly any views. But yea we could use that channel to put out some more community made videos for sure. Looking at what's on there now, a couple case review videos got decent views but most of them are just way too specific and boring.

Instead of : Started by xxxxx Started on xxxxxxx Could be: Started by xxxxx on xxxxx Or Started on xxxxx by xxxxxx

I think ehw needs a new YouTube channel separate from the podcast channel. I'm not sure how this would work exactly as far as uploading, but if there were just a handful of interesting videos of various PC stuff, it would go a long way towards creating a presence there that might bring in a few users. Maybe there could be a cloud storage that people could dump video files onto that could later be uploaded to an ehw channel, assuming anyone would want to donate some video content. I'd be happy to put a few in game benchmark videos or just gameplay videos up. There are definitely some unique builds here, and a video that was something like "demanding game running on 4090 SLI" would definitely bring in some views.

Personally I wouldn't mind each line being chunky as long as longer titles can be displayed, but if the "started by" part had to go in order to make the titles more visible, that would be a worth while trade off.

What is your top game pick of 2023, and why? My pick: Armored Core 6 This game was so overdue for me as someone who absolutely loved the AC series on PS2, where it shined the most. From Soft did a great job of recapturing the nostalgic feel while also making the game feel modern enough to avoid feeling dated. The graphics seem simple at first as you inspect the environments. It all feels a little low-fi initially, but once the real mech battles ramp up and you see how many particle effects are being pushed, it all clicks into modern mech perfection. My only complaint is that it's a seriously hard game. I am yet to beat it, stuck on a late game boss. But the game gives so many options for how to spec your mech, it's just a matter of time until I find the right combo. I wish I had played more 2023 games, but 2023 has felt like year of build up for the high-quality games that big studios have yet to bring to fruition. Here's hoping 2024 is the year UE5 titles, and titles from competing engines really start to shine. Also, most anticipated game for 2024? Mine: HellBlade.

Ever since dedicated graphics cards can into existence, the trend has been to make them bigger, use more power, and create more heat. That doesn't mean they haven't also gotten a lot more efficient though. A GeForce 4060 uses ~115w and is vastly more efficient that a GeForce 8800. This stuff has a lot more to do with physics and the physical limitations of what's possible than any kind of design choices. As we get closer and closer to the atomic limit of transistors, something has to give. Sure there is always a greed factor, but by far the thing driving GPUs to be more expensive and less of a raw power increase gen to gen is physics. That's a big part of the reason why things like AI upscaling and other AI rendering techniques will be more and more important. It's a way to keep increasing visuals without relying purely on more and more transistors. Already what we call 3 or 4 or 5mn nodes really just means some parts of a chip are that small, but there are a bunch of other parts of chips that never got past the 16nm barrier and are still fabricated at 16nm, for example. Simply put, as the physics starts to get more and more in the way of advancement, the chips just have to get smarter.

No idea on the specifics of their dies, I just know there is a huge gap between the 7800XT and 7900XT.

As it is right now, the 7900XT is typically faster than everything Nvidia has aside from the 4080 and 4090 in normal games and RT games but falls short on PT. I am guessing the 4070Ti Super will slot between the 7900XT and 7900XTX. No doubt AMD will respond with price cuts, and I think we might get to see some much needed competition on prices. If the 4070Ti Super sells anywhere near MSRP, AMD prices will need to come down a lot. Right now the 7900XTX is going for just under $1000, that's going to have to come down to ~$750 to compete with the 4070Ti Super. I think AMD could also add some cards to their stack. I doubt they will release anything higher end than a 7900XTX, but they just announced the 7600XTX, they could do a 7700XTX and more importantly a 7800XTX. There is a wide gap in terms of both price and performance between the 7800XT and 7900XT. A 7800XTX would fill a huge gap from AMD and put that much more pressure on Nvidia. I don't think AMD has much headroom to go above a 7900XTX due to power consumption, but if they could squeeze in a 7950XT/XTX and sell that for $900 and drop the 7900XTX to $800 or less, that would be pretty sweet.

Just a thought but you could sell the cpu you were going to give your buddy and then just give him the money instead so he could go with DDR5.

Maybe a quick 15 second YouTube video commercial? Then we could link it to stuff. Maybe YouTubes algorithm would recommend it to some tekkies with a little luck. "Are you tired of those other forums?" **cuts to clip of Linus apologizing for something** "Try EHW!"

AMD's stock has gone up 21.5% in the last month. Seems like Investors are anticipating some real competition in AI from AMD. Nvidia's stock is also up by 11% over the last month. Intel's stock is up 15.3% over the last month. Could be the start of a new revolution in AI hardware. Once we have all three of these companies consistently able to compete with each other with good AI hardware, it's going to be very interesting times.

Why do you need that?

https://www.techspot.com/news/101435-amd-latest-radeon-rx-7600-xt-double-vram.html If AMD launched a few more cards like this, they might have a good answer to the Super cards.

Less than $300? That seems pretty high. $150 should be enough for a good Intel board. Just avoid ASRock.

I think it's highly likely that AMD will lower prices.

Why not use a drip if you're using a kettle?

How do you guys determine the used prices? Do you guys usually sell your GPUs on Ebay? I feel like it is absolutely essential to be effectively upgrading when you are in the high end.

haha yea I guess that makes sense.

Hmm, I never would have guessed that. I wonder why that is.

KInda a side note that your comment made me think about, what do you guys usually do with your old GPU when you upgrade? For a long time now I have been in the habit of selling my old GPU right after I get a new one, so the price to upgrade is only ever as high as the difference between what I can sell my old one for and what the new one costs. Because of Covid/mining, when I sold my 1080Ti and upgraded to an RX6800, I actually made a little bit of money (like $10). After that I sold my RX6800 for $400 and bought the 7900XT for $800. Do you guys do this sort of thing?

Instant coffee? At home? Instant coffee is for camping, isn't it? I have a cold brew maker, which is really just a pitcher with a coffee filter in it, so I drink cold brew at home. Supposedly cold brew is less acidic than hot coffee.

You can apparently still download and install SteamOS: SteamOS (steampowered.com) These devices are just small AMD (usually) PCs with laptop CPUs and GPUs. Windows works normally on them. No idea if other manufacturers would be able to just ship SteamOS devices if they wanted to, but I don't think of it as a real selling point. Valve has refined SteamOS on the Steam Deck to an impressive degree, but it's still a long way off from Windows.

I agree, definitely not an innovative game. I think steam awards go out based on total players and playtime to some degree, and a lot of people played Starfield for a good amount of time before realizing it was pretty disappointing.