Hardware discussion spun off M52 prediction thread - Page 2

Uncwilly · 2021-05-09, 00:16

Quote:

Originally Posted by drkirkby

I think rule 0 should be that all rules are printed in the same size text, so rules that disqualifies people are not in small text. See screenshot from my mobile phone.

The problem is on your end. There are no tags in the first post to change the point size of any text.

Xyzzy · 2021-05-09, 00:57

A 2080Ti/3060Ti/3070/3080/3090 will average from 60 to 120 MH, So using Nicehash you can "earn" from $3 to $15 a day.

We mined for a few weeks in January and now we have ~$450 in BTC.

While we really like the idea of anonymous and decentralized "currency", we dislike the fact that mining, in general, is really bad for the environment.

tuckerkao · 2021-05-09, 02:14

Quote:

Originally Posted by Uncwilly

The problem is on your end. There are no tags in the first post to change the point size of any text.

I can read those texts of the rules just fine. It's the best to view the forum using a PC than the smartphone. The screen resolution has to be wide enough for all the texts to display properly. I believe it's possible to switch the smartphone view to horizontal instead of vertical.

tuckerkao · 2021-05-09, 02:23

Quote:

Originally Posted by Viliam Furik

Nope. I think I already made my components known in some previous thread, but if not, I have a Ryzen 9 3900X, RTX 2080Ti, and a Radeon VII. The GPUs are now however mining Ethereum (because unfortunately, money is always needed).

I've seen your GPU trial factoring up to 2^80 several times, so I'd like to know the amount of hours required to factor M168,xxx,323 from 2^78 to 2^80 using your hardware. You don't have to do my assignments. I need those statistics because if I ever want a new graphic card, I'd like it to be able to cover up to 2^82 with the trial factoring of 2^81 to 2^82 of M462,xxx,xxx not longer than a full day.

Quote:

Originally Posted by drkirkby

Can you make enough money to actually pay for the graphics card? Or is it a case of you bought a graphics card for playing games, and it earns you a little money. Clearly if it earns you $25/month, it would take an awful long time to recover the cost of the graphics card.

Dave

I play simulation games like "Cities: Skylines", just don't run mfaktc during the meantime, so purchasing a newer machine with a high-end GPU will definitely speed up everything. It would be the impatient waits if I have to check out my Prime95 all the time.

I usually just have my father paying for the CPU and GPU as long as I can show that I'm able to accomplish the exponent tasks I've promised, only time will tell.

Quote:

Originally Posted by Xyzzy

A 2080Ti/3060Ti/3070/3080/3090 will average from 60 to 120 MH, So using Nicehash you can "earn" from $3 to $15 a day.

We mined for a few weeks in January and now we have ~$450 in BTC.

Mining for a Mersenne Prime seems to be a more fun job especially when UncWilly shows up all different types of the emo icons.

drkirkby · 2021-05-09, 05:25

There are 11 HTML errors on the page with the rules

https://validator.w3.org/check?uri=h...Inline&group=0

so the browser has to take a guess at how to render the page. The results are undefined.

The homepage for the forum has no errors

https://validator.w3.org/check?uri=h...org%2Fservices

unlike the Primenet web site, where the homepage has 41 errors
https://validator.w3.org/check?uri=h...Inline&group=0

The menu structure of that is invisible in Firefox on my iPhone, although it works fine on the Safari and Google browsers.

Although the point was made by tuckerkao that the site is best viewed with a desktop, not an phone, it is worth noting that Google are saying the majority of users coming to a site are likely to be using a mobile device

https://developers.google.com/search/mobile-sites

So to quote them, with the emphasis in bold on the Google website

"Mobile is changing the world. Today, everyone has smartphones with them, constantly communicating and looking for information. In many countries, the number of smartphones has surpassed the number of personal computers; having a mobile-friendly website has become a critical part of having an online presence. If you haven't made your website mobile-friendly, you should. The majority of users coming to your site are likely to be using a mobile device."

paulunderwood · 2021-05-09, 07:50

Not really anything to do with M52 predictions... If you want a site to be "mobile friendly" why not do the changes needed yourself? Or pay somebody else for the makeover? Why do you expect things to be done to suit your little screen, for free? Perhaps you could approach Google to do the work -- they can afford it!

tuckerkao · 2021-05-09, 08:19

Quote:

Originally Posted by paulunderwood

Not really anything to do with M52 predictions... If you want a site to be "mobile friendly" why not do the changes needed yourself? Or pay somebody else for the makeover? Why do you expect things to be done to suit your little screen, for free? Perhaps you could approach Google to do the work -- they can afford it!

I think UncWilly will consider the technical issues of the mobile device as part of the hardware discussion which he had already created a separated spunning thread, but drkirkby kept posting on the original M52 Prediction thread.

kriesel · 2021-05-09, 17:52

(Mods might wish to split html standards compliance into a separate thread, perhaps placing it in the forum feedback subforum)

A bit of devil's advocate here for both sides, after 3 years of reference blog, 4 years of mersenneforum use, & preceded by 20+ years of direct or indirect responsibility for a major university's small department's web site & server from scratch. (That small department provided a few megadollars annually of products and services to governments and businesses worldwide, and occasionally for use for decades at or under the South Pole, undersea or in low earth orbit or other harsh unforgiving environments.)

On the one hand:
The Mersenne forum software is a layer interposed between us users and whatever html is emitted. It apparently subsets the possibilities significantly. It sets limits on what a user can do. Neither I nor Xyzzy know of a way to do the equivalent of html anchors inserted in a blog post. It would be very useful for combining functions of https://mersenneforum.org/showpost.p...22&postcount=1 and https://mersenneforum.org/showpost.p...83&postcount=2 for more user convenience.
If one experiments with putting certain html constructs in a forum post, as I have recently, it is just text that displays, not having the intended effect, in the limited testing I tried.

On the other hand:
At the departmental web site I started, every page creation or edit went through numerous checks while I was hands-on with it.
Among them were technical accuracy, stale/current content, tone, style, ambiguity, grammar, spell check, dual metric/American units, ADA (Americans with Disabilities Act) compliance, search keywords, robots enable or deny, tolerance of variable browser window sizing, top-3 browsers compliance, and html validation. (Some html code generators' output BADLY failed validation; notably MS Frontpage ~20-25 years ago. And browsers didn't always behave as expected.) The ADA compliance was a mandatory condition associated with the university's receipt of ~$10⁹ US R&D funding annually, via various US federal grants mostly, so ADA compliance was a big deal to top university management.
After the hands-on duties passed to a subordinate, compatibility with a greater variety of screen sizes from 27" CAD monitor to laptop to tablet to phone got incorporated in the checklist. The site was switched from direct html editing to a CMS (content management system), which had pros and cons. Fixing a validation issue became more difficult, as I recall.

In practice:
There's a validation error with https://mersenneforum.org/showpost.p...&postcount=215 which is about validation errors.
I wonder, how does anyone propose to fix that post's compliance? https://validator.w3.org/check?uri=h...org%2Fservices

This is not a facetious or rhetorical question. It's likely my every reference blog post fails validation; probably every forum post.
The forum software's html generator software probably has multiple standards compliance bugs.
https://mersenneforum.org/showthread...983#post577983 fails validation.
Try finding the errors in Firefox in view-source:https://mersenneforum.org/showthread...983#post577983 which is over 2100 lines of output.
Then try to find a way to fix them, from within the forum environment as it is.
In that post, I am not using centering. A tag with center is not present in my 410 plain text lines of input (checked with a computer search of the input text). The end tag for center omitted ... error seems to be coming from what the forum software generates from my input. That post, and this post, were composed on a 17" laptop.

drkirkby · 2021-05-26, 06:06

Quote:

Originally Posted by moebius

No mfaktc uses too much cpu resources, but you can run gpuowl (PRP/LL or P-1)and prime95 at the same time with maybe 1-5% slowdown on (at least on a Vega64 or Radeon VII).

My experience is the dead opposite of yours. I find that mfaktc uses virtually no CPU time on my system Linux system, but gnuowl was using 100% of one core. I have 52-cores, so having one core tied up is not so serious, but I stopped using gnuowl as my graphics card is not well suited for GIMPS.

I'm using

* Dell 7920 with 2 x Xeon 8167M 26-core 2.0 GHz CPUs, with 35 MB cache on each CPU
* 384 GB RAM.
* Nvidia Quadro P2200 graphics card.

Uncwilly · 2021-05-26, 14:07

Quote:

Originally Posted by drkirkby

but I stopped using gnuowl as my graphics card is not well suited for GIMPS.

* Nvidia Quadro P2200 graphics card.

Aren't you able to get around 380 GHz-days/day on that?
If I had that at my disposal, I would be overjoyed.

axn · 2021-05-26, 14:31

Quote:

Originally Posted by Uncwilly

Aren't you able to get around 380 GHz-days/day on that?
If I had that at my disposal, I would be overjoyed.

380 Gd/d of TF is not that impressive. Even a lowly 1650 would get you ~1000. On gpuowl, I think it gets something like 40 Gd/d?

2021-05-09, 05:25	#16
drkirkby "David Kirkby" Jan 2021 Althorne, Essex, UK 458₁₀ Posts	There are 11 HTML errors on the page with the rules https://validator.w3.org/check?uri=h...Inline&group=0 so the browser has to take a guess at how to render the page. The results are undefined. The homepage for the forum has no errors https://validator.w3.org/check?uri=h...org%2Fservices unlike the Primenet web site, where the homepage has 41 errors https://validator.w3.org/check?uri=h...Inline&group=0 The menu structure of that is invisible in Firefox on my iPhone, although it works fine on the Safari and Google browsers. Although the point was made by tuckerkao that the site is best viewed with a desktop, not an phone, it is worth noting that Google are saying the majority of users coming to a site are likely to be using a mobile device https://developers.google.com/search/mobile-sites So to quote them, with the emphasis in bold on the Google website "Mobile is changing the world. Today, everyone has smartphones with them, constantly communicating and looking for information. In many countries, the number of smartphones has surpassed the number of personal computers; having a mobile-friendly website has become a critical part of having an online presence. If you haven't made your website mobile-friendly, you should. The majority of users coming to your site are likely to be using a mobile device." Last fiddled with by drkirkby on 2021-05-09 at 05:28

2021-05-09, 07:50	#17
paulunderwood Sep 2002 Database er0rr 5·937 Posts	Not really anything to do with M52 predictions... If you want a site to be "mobile friendly" why not do the changes needed yourself? Or pay somebody else for the makeover? Why do you expect things to be done to suit your little screen, for free? Perhaps you could approach Google to do the work -- they can afford it! Last fiddled with by paulunderwood on 2021-05-09 at 08:11

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
2020 the year in space prediction thread.	Uncwilly	Science & Technology	14	2021-05-13 07:17
General Hardware Discussion	veganjoy	jvang	21	2020-10-31 14:20
Do-it-yourself, crank, mersenne prediction thread.	Uncwilly	Miscellaneous Math	85	2017-12-10 16:03
2045 eclipse/future prediction thread	MooMoo2	Lounge	2	2017-08-26 23:39
Old Hardware Thread	E_tron	Hardware	0	2004-06-18 03:32

2021-05-09, 00:57	#13
Xyzzy Aug 2002 21D3₁₆ Posts	A 2080Ti/3060Ti/3070/3080/3090 will average from 60 to 120 MH, So using Nicehash you can "earn" from $3 to $15 a day. We mined for a few weeks in January and now we have ~$450 in BTC. While we really like the idea of anonymous and decentralized "currency", we dislike the fact that mining, in general, is really bad for the environment.

2021-05-09, 17:52	#19
kriesel "TF79LL86GIMPS96gpu17" Mar 2017 US midwest 2⁴·3·163 Posts	(Mods might wish to split html standards compliance into a separate thread, perhaps placing it in the forum feedback subforum) A bit of devil's advocate here for both sides, after 3 years of reference blog, 4 years of mersenneforum use, & preceded by 20+ years of direct or indirect responsibility for a major university's small department's web site & server from scratch. (That small department provided a few megadollars annually of products and services to governments and businesses worldwide, and occasionally for use for decades at or under the South Pole, undersea or in low earth orbit or other harsh unforgiving environments.) On the one hand: The Mersenne forum software is a layer interposed between us users and whatever html is emitted. It apparently subsets the possibilities significantly. It sets limits on what a user can do. Neither I nor Xyzzy know of a way to do the equivalent of html anchors inserted in a blog post. It would be very useful for combining functions of https://mersenneforum.org/showpost.p...22&postcount=1 and https://mersenneforum.org/showpost.p...83&postcount=2 for more user convenience. If one experiments with putting certain html constructs in a forum post, as I have recently, it is just text that displays, not having the intended effect, in the limited testing I tried. On the other hand: At the departmental web site I started, every page creation or edit went through numerous checks while I was hands-on with it. Among them were technical accuracy, stale/current content, tone, style, ambiguity, grammar, spell check, dual metric/American units, ADA (Americans with Disabilities Act) compliance, search keywords, robots enable or deny, tolerance of variable browser window sizing, top-3 browsers compliance, and html validation. (Some html code generators' output BADLY failed validation; notably MS Frontpage ~20-25 years ago. And browsers didn't always behave as expected.) The ADA compliance was a mandatory condition associated with the university's receipt of ~$10⁹ US R&D funding annually, via various US federal grants mostly, so ADA compliance was a big deal to top university management. After the hands-on duties passed to a subordinate, compatibility with a greater variety of screen sizes from 27" CAD monitor to laptop to tablet to phone got incorporated in the checklist. The site was switched from direct html editing to a CMS (content management system), which had pros and cons. Fixing a validation issue became more difficult, as I recall. In practice: There's a validation error with https://mersenneforum.org/showpost.p...&postcount=215 which is about validation errors. I wonder, how does anyone propose to fix that post's compliance? https://validator.w3.org/check?uri=h...org%2Fservices This is not a facetious or rhetorical question. It's likely my every reference blog post fails validation; probably every forum post. The forum software's html generator software probably has multiple standards compliance bugs. https://mersenneforum.org/showthread...983#post577983 fails validation. Try finding the errors in Firefox in view-source:https://mersenneforum.org/showthread...983#post577983 which is over 2100 lines of output. Then try to find a way to fix them, from within the forum environment as it is. In that post, I am not using centering. A tag with center is not present in my 410 plain text lines of input (checked with a computer search of the input text). The end tag for center omitted ... error seems to be coming from what the forum software generates from my input. That post, and this post, were composed on a 17" laptop.