add arrow-down arrow-left arrow-right arrow-up authorcheckmark clipboard combo comment delete discord dots drag-handle dropdown-arrow errorfacebook history inbox instagram issuelink lock markup-bbcode markup-html markup-pcpp markup-cyclingbuilder markup-plain-text markup-reddit menu pin radio-button save search settings share star-empty star-full star-half switch successtag twitch twitter user warningwattage weight youtube

PC Part Picker needs to update its selection list of parts

minervyx
  • 30 months ago

1) Categorically remove the parts on the list which either (a) have a review average of under 3 stars or (b) are over 3 years old and have no reviews.

2) For all CPU's and GPU's that are several years old, make note of their age and their original price. A $200 CPU with 4.5 average and 50 ratings sounds good until you realize that it's 4 years old.

3) Add more cases and more high-resolution images of the cases in their pages. There are particularly a lack of small form factor cases and high-resolution monitors.

4) Go for a quality over quantity approach. For example, only have ~40 of each part (instead of 200), but every part should be of suitable quality for the target consumer and well-explained in its functionality.

Comments

  • 30 months ago
  • 10 points

Some thoughts (also posted in the Discord server):

  • We don't want to be in the business of telling people what they should or shouldn't be buying by removing options.
  • People, even folks who have low-reliability PSUs, still need to be able to model their systems.
  • Our goal is not to tell people what to buy, but to provide enough info in easily consumable ways, to make the decisions themselves (i.e. benchmarks, testing). Benchmarking and testing are coming, FWIW.

EDIT

  • We're gradually working on images. It takes time. High resolution images are surprisingly hard to get hold of for things that aren't new product releases and aren't the high end items. manirelli and ThoughtA recently did a pass through all our products revisiting all the images, repicking and sorting them. You're talking tens of thousands of products. They didn't quit, so we're good now.
  • 30 months ago
  • 0 points

We don't want to be in the business of telling people what they should or shouldn't be buying by removing options.

With the logic you provided of "less options = more discouragement", the act of not adding more options is also discouraging.

What makes X number of parts inherently any better of a number than X +/- Y number of parts?

  • 30 months ago
  • 2 points

We're working on adding more parts. If there are parts we're missing that you'd like to incorporate into builds please let us know and we'll be happy to add them in.

  • 30 months ago
  • 1 point

I appreciate your open-mindedness. However, this is only a bandaid for the problem. My suggestions alone nor a handful of a few other voluntaries with ample time on their hands aren't going to get the selection of parts to what it needs to be.

There needs to be a very quick accessible way to make suggestions on the parts page (for example, "Is the part you're looking for not on this list? Click here to type a suggestion").

  • 30 months ago
  • 3 points

Looking through your submitted topics, I don't see that you've ever used our part additions forum. If there is a part you'd like to realistically use, you can suggest it there and we'd be happy to add it if it fits in our existing categories.

But if you're unhappy with the current selection - I get it. We're actively working on adding parts and reducing latency on new part introductions. It's a time intensive process, but we're doing what we can. In the last month we've identified areas to improve and we're working on it.

But regarding removing unfavorable parts per your original topic, we are not going to do that.

  • 30 months ago
  • 1 point

Fair enough. I will submit additions.

  • 30 months ago
  • 1 point

1) Most of the reviews for parts have very few reviews, and is by no means a statistically significant indicator of how reliable/well-made a product is.

2) This site exists to help you with putting together parts and making sure they are compatible with one another. The forums, Discord, and other communities (such as Reddit's Build a PC) exist to tell you whether or not said combination of parts is an effective use of your money for your usage.

3) The images are not manually added, they are scraped from the sites of retailers who list them (IIRC). As a result if you have a problem with them you could take them up with the retailers. You have to realize there are hundreds of thousands (if not more) parts that this site conveniently organizes, categories, and provides information on. From there you have links and the information to find more. For them to add high resolution photos for every single product (making sure they are allowed to use said photos) is a herculean undertaking and by no means an effective use of their time.

4) The target consumer is anyone building a computer, there are not 40 parts of any category that will necessarily adequately provide for any usage scenario. Instead the focus is to provide you a snapshot of everything that is available at the listed major etailers/retailers and what they are compatible with. A less is more approach is absolutely counter to the core of this site.

  • 30 months ago
  • 1 point

The point of the site is to give people a list of parts, not tell them which ones are best.

1) Say I have some old CPU that barely sold, but I still want to have it in a list to check for motherboard compatibility. Why should I not have the choice of creating a list identical to my system, then doing things such as running it through forumns for suggestions?

2) That makes no sense. Just read that it says "i5-2400" and assume its older than something that says "i5-7400". You don't need to list them.

3) There aren't many SFF cases to begin with (not when compared to ATX ones). I'll agree that there should be more high-res and high-end monitors (specifically 3440x1440 ones).

4) This site's job isn't to write an essay on every part. People can click on a CPU, get its name, and look up one of hundreds of reviews on it. If the reviews are dated 2011, then people know its an outdated part and will look for newer versions.

  • 30 months ago
  • 1 point

Why should I not have the choice of creating a list identical to my system, then doing things such as running it through forumns for suggestions?

How about an option for people who are trying to build a new system which blacklists parts which wouldn't be relevant to them?

2) That makes no sense. Just read that it says "i5-2400" and assume its older than something that says "i5-7400". You don't need to list them.

Beginners aren't aware of this. At least have some way of beginners to know this. For example, writing a sentence above the table of CPU search results such as "For Intel CPU's, the first of the 4 numbers after the hyphen refers to the generation number. (i.e. an i_-6XXX is more recent than an i_-4XXX)

3) There aren't many SFF cases to begin with (not when compared to ATX ones).

Not if we're using PCPP's search engine.

4) This site's job isn't to write an essay on every part.

That's a hyperbolic statement. It doesn't have to be all-or-nothing. The site can be more informative without needing to write a review on each part.

More tags, for example, so users can search more effectively. For example, a tag for the product's year of release (or the "era" the product was released in).

For the information on the initial table, it's really silly to just have (for example) the clock speed and number of cores. Because a X GHz x. Y-cores from 2013 is a lot different than one from 2017.

  • 30 months ago
  • 2 points

For the information on the initial table, it's really silly to just have (for example) the clock speed and number of cores. Because a X GHz x. Y-cores from 2013 is a lot different than one from 2017.

We want to be data driven and not editorial (as much as we can). That's why we're pursuing benchmarking on a scale no one has seen in PC tech publishing to date. It's coming, soon.

  • 30 months ago
  • 1 point

This would be very helpful for CPU's and GPU's. I hope it works out.

  • 30 months ago
  • 2 points

We've invested a stupid amount of money into it - I hope it works out too. :)

  • 30 months ago
  • 1 point

How does PCPP make money? I'm not bombarded with ads or anything

  • 30 months ago
  • 1 point

It already blacklists parts that aren't compatible. So what if they choose a 4790k instead of a 7700k? Someone on the forumns will tell them, and they can fix it.

You are asking PCPP to write a Techquicke about everything.

The bottom-line) PCPP is a tool people use to pick parts, while the rest of the internet is where they can get recommendations.

  • 30 months ago
  • 1 point

You are asking PCPP to write a Techquicke about everything.

No. Moreso, tags and a few tips.

  • 30 months ago
  • 1 point

Essentially a techquickie. And on tens of thousands of products, thats a ton of work. Plus, the point of the site isn't to give "tips" on products. Thats like making a powerpoint and having pop=ups saying "A black slide with white text would look better, so you should pick that".

  • 30 months ago
  • 1 point

Information is provided about clock speeds and such anyway. I'm not at all suggesting the site should play favorites in the market or tell people which products to get. However, just general indications. (i.e. "this CPU is 5+ years old. it may be obsolete" IS not unlike "you may need to update your BIOS to accomodate the CPU"

[comment deleted by staff]
  • 30 months ago
  • 1 point

Please keep it polite.

[comment deleted]
[comment deleted by staff]
  • 30 months ago
  • 0 points

Why not have 1000 parts for each category then? Why not add a bunch of parts from 2008 or ones that don't work well?

What standards are employed apart from keeping the completely arbitrary choices on list exactly the same out of tradition?

  • 30 months ago
  • 3 points

2008? The site was founded in 2010. There realistically aren't many people updating a 2008-era system to 2017 and retaining parts. A five year spread is at most what we see.

  • 30 months ago
  • 2 points

Its easily said than done... Funny how ppl complain and don't see the hard work to get this site where it is now. Why doesn't he create his own site to his desired specifications.. Didn't think so. :)

  • 30 months ago
  • 0 points

Would it be economical to tag CPU's/GPU's by their year of release?

  • 30 months ago
  • 5 points

If you have performance numbers (benchmarking) and prices, is year of release relevant?

  • 30 months ago
  • 2 points

Well that can be a tricky area. Year of release can play a role in things like DX10/11/12 or other things that are more common to a certain time period. Not that I am saying going by the year is a good or perfect gauge for this, but it is just a point in which year could be helpful. I would rather see more specificity on specs, however, rather than release year. GJ with the site btw. :)

  • 30 months ago
  • 1 point

^^

  • 30 months ago
  • 1 point

Except you are requesting they arbitrarily determine and make available only good parts from bad ones...which is a logical fallacy of your argument.

Sort

add arrow-down arrow-left arrow-right arrow-up authorcheckmark clipboard combo comment delete discord dots drag-handle dropdown-arrow errorfacebook history inbox instagram issuelink lock markup-bbcode markup-html markup-pcpp markup-cyclingbuilder markup-plain-text markup-reddit menu pin radio-button save search settings share star-empty star-full star-half switch successtag twitch twitter user warningwattage weight youtube