This video by Hardware Haven, titled “6 Tips for Finding Good Deals on PC Hardware,” provides actionable strategies for finding cheap and heavily discounted used computer components, servers, and office PCs.
Here is a breakdown of the six main tips covered in the video:
Find local ITAD (IT Asset Disposition) businesses [01:40]
These companies wipe, recycle, and resell old equipment decommissioned by businesses. Since they deal in bulk, you can often score incredible deals on enterprise or office hardware. If they don’t have a public storefront, you can sometimes find them by searching for local eBay sellers dealing in bulk office PCs.
Use eBay local pickup [04:27]
For large or heavy items like servers and tower systems, sellers bake the high cost of shipping directly into their asking price. By filtering for “local pickup only,” you avoid those shipping costs and can often secure a much lower base price.
Get creative with the platforms you use [07:46]
Don’t rely exclusively on eBay. Check out niche component sites like jawa.gg, local thrift stores, university surplus auctions, or GovDeals. When using Facebook Marketplace, browse generic local buy/sell groups rather than PC-specific ones—you are more likely to find sellers who don’t know the exact value of their hardware.
Use unconventional search terms [11:04]
Instead of searching for highly sought-after product lines, search for the specific, less-common processors inside them. You can also find steep discounts by intentionally searching for minor cosmetic defects using words like “missing,” “cracked,” or “damaged.” For local marketplaces, try generic, non-technical terms like “CPU tower” or “good computer.”
Put in more effort than other buyers [13:46]
You can find extreme discounts if you’re willing to buy hardware that requires a bit of troubleshooting. This includes buying locked systems and clearing the BIOS, flashing custom firmware onto cheap Chrome boxes, or taking a calculated risk on “for parts” listings that might just need something as simple as a new CMOS battery.
Be patient [18:03]
Used enterprise gear hits the secondary market in massive waves when companies do their hardware refresh cycles. If a particular component is overpriced right now, don’t overpay—wait for the next wave of surplus gear to flood the market and drive prices back down.
Thanks for admitting it’s done with AI, I’ll wait for someone else to provide humanely useful information since presumably about 60% of that text is hallucinated.
It’s definitely not 60% hallucinated, and that seems like an insane assumption to me, but I’m new here and did not realize that use of AI was considered inhumane.
thank you for being honest about it but please remember that each and every AI query is absolutely devastating to the environment!! It’s pretty easy to just type up a quick summary (or let someone else do it) rather than do that damage
The usage of commercial AI’s hosted in data centers encourages the growth of that industry with the construction of more data centers, which are hazardous to the populations that live around them, encourage pretty severe water waste in areas with already dwindling groundwater, and collectively use a tremendous amount of particularly polluting electricity since the local grids cannot adequately supply them, meaning many resort to using inefficient portable power generation on-site, and in many cases is causing energy prices to spike.
While I personally think that AI use is virtually always detrimental (since there is growing evidence that AI usage appears to cause people to atrophy their critical thinking skills, and many deferring to incorrect answers), for those that still insist on using it, using a self-hosted local LLM is the most ethical option, as it does not encourage the construction of more data centers, and avoids feeding multinational fascist corporations with more data.
No, but it’s not the energy use I’m talking about (though I’ll do some research on video games now, there’s always more for me to learn!) - I’m referring to the water usage. The average ChatGPT session (with multiple queries) is the equivalent of destroying roughly a bottle of water. But if we are thinking about power, data centers also run diesel powered generators when there’s a high demand for queries which spew out pollution. And then its worth considering that each data center as a whole is pretty awful, and if there were less demand for them, i.e. people not using genAI, the CEOs would be motivated to make them less shitty… every query shows them that, yeah, this IS a great idea and they should build more data centers :/
I don’t mean to like make you feel bad for making a quick tl;dr using AI, it’s just that this is pretty important to me! We all have to pick and choose our battles, but of all the things we do in our day-to-day lives that impact the environment and the world around us, typing up our own summary instead of using AI seems like a really easy option to me.
Here you go. (Disclaimer: I used AI)
This video by Hardware Haven, titled “6 Tips for Finding Good Deals on PC Hardware,” provides actionable strategies for finding cheap and heavily discounted used computer components, servers, and office PCs.
Here is a breakdown of the six main tips covered in the video:
Thanks for admitting it’s done with AI, I’ll wait for someone else to provide humanely useful information since presumably about 60% of that text is hallucinated.
It’s definitely not 60% hallucinated, and that seems like an insane assumption to me, but I’m new here and did not realize that use of AI was considered inhumane.
thank you for being honest about it but please remember that each and every AI query is absolutely devastating to the environment!! It’s pretty easy to just type up a quick summary (or let someone else do it) rather than do that damage
That is not accurate. My AI query used far less energy than the average gamer uses in a minute of gameplay.
I’m not a gamer, but I’m curious – do you hold the same attitude towards video games?
The usage of commercial AI’s hosted in data centers encourages the growth of that industry with the construction of more data centers, which are hazardous to the populations that live around them, encourage pretty severe water waste in areas with already dwindling groundwater, and collectively use a tremendous amount of particularly polluting electricity since the local grids cannot adequately supply them, meaning many resort to using inefficient portable power generation on-site, and in many cases is causing energy prices to spike.
While I personally think that AI use is virtually always detrimental (since there is growing evidence that AI usage appears to cause people to atrophy their critical thinking skills, and many deferring to incorrect answers), for those that still insist on using it, using a self-hosted local LLM is the most ethical option, as it does not encourage the construction of more data centers, and avoids feeding multinational fascist corporations with more data.
No, but it’s not the energy use I’m talking about (though I’ll do some research on video games now, there’s always more for me to learn!) - I’m referring to the water usage. The average ChatGPT session (with multiple queries) is the equivalent of destroying roughly a bottle of water. But if we are thinking about power, data centers also run diesel powered generators when there’s a high demand for queries which spew out pollution. And then its worth considering that each data center as a whole is pretty awful, and if there were less demand for them, i.e. people not using genAI, the CEOs would be motivated to make them less shitty… every query shows them that, yeah, this IS a great idea and they should build more data centers :/
I don’t mean to like make you feel bad for making a quick tl;dr using AI, it’s just that this is pretty important to me! We all have to pick and choose our battles, but of all the things we do in our day-to-day lives that impact the environment and the world around us, typing up our own summary instead of using AI seems like a really easy option to me.