Jump to content

Recommended Posts

Posted

I hope that IPS becomes faster this year as long as speed is more relevant each day as a ranking factor.

Maybe parts of IPS could be loaded by the API and JavaScript, following the model used by Instagram and Twitter.

The basic HTML structure loads fast and the contents are lazy loaded on demand (on view port) by the API.

New websites are adopting Next.js for fast loading.

I think it’s time for a huge IPS modernization on terms of speed. The basic tools are already here.

Posted

I usually put a lot of effort on website optimization, and I certainly would like to see IPS paying more attention to performance.

But looking at results from other popular websites such as Youtube, Facebook or Amazon, the values are not that different.  I checked a few other landing pages from Google, with mostly static content, and the results were even worse.  So, I guess IPS is not that bad.

Posted

I will admit it has been super huge challenge to gain the good ranking of the core web vitals. I've been pulling out stuff left and right ditch excess graphics, functions, etc. Slowly gaining a bit of ground and Google showed that I just only needed a bit more improvement. Couple days later Google now took all the dekstop pages and marked them as poor again. Really tough to make a perfect score even close being my biggest problem is Google Adsense and the CLS problem. Then one day I could have wonderful timing now its telling me LCP is now too slow at >2.5 seconds again. 

Posted (edited)

Like myself, I managed to pull it off... Marginally but did it... I've expanded more ads on the thread listing page and the forum post pages too. Still managed to squeeze through barely making it all fit. Ugh!

127975732_Screenshotfrom2021-02-1007-25-20.thumb.png.c79abdb64b772b42c071b6213b319ef0.png

I'll use any web based test tool from the outside. I can't really use my browser developer tools being my local internet here is slow. So the scores are always a touch lower and never quite reach the mark ever. Now web based test like GT Metrix works well being it like any customer coming to the site.

 

 

Edited by Mopar1973Man
Posted

Still waiting for the Google to retest all this. I've managed to knock the LCP (Largest Colorful Paint) to just a mere 421 URL left. I'm in the process of backing up all the attachments and now going to recompress all my attachments and photos. I found a GUI style picture file compressor called "Trimage" then there is a terminal based called "jpegoptim" that work directly on your server terminal. I'm going to do most of the processing at home and re-upload the files again this way I don't tank the server load with monster recompress of 8 GB of files. Currently on the desktop side the only poor I got is CLS (Content Layout Shift) again waiting for Google to retest this out. I'm my testing I managed to reach 0.08 on most pages. 

The biggest is the CLS and diving into the CSS and do a search for "google" it will pop up. Now you can set the amount of space in the CSS to fix the CLS or at least reduce it. I aimed for reduction mostly. Kind of hard with dynamic layout design like Invision.

Timing is basically getting your server optimized as best you can then start drop out the drag. More or less how I build diesel truck for customers. Lose as much drag and weight then start adding power to the engine. This way your not adding power to huge archor or excessive weight and trying to drag the weight. (Switch back to web) More or less ditch the heavy graphics as much as possible, go more "Plain Jane" now you can add back in the goodies and watch till Google whines about either content shift or timing. 

Maybe that's why I drive a 500 HP Diesel Truck that most just think is stock...
https://mopar1973man.com/storage/attachments/monthly_2020_06/15915332979951430079351958718559.jpg.3c0f371c8c1300491ef55ed800e1ea0c.jpg

Posted

That's cheating... Yeah IPS might have a zero CLS shift because they do not have any Google Adsense being displayed. Where most of us use the Google adsense for income. Another thing to look at is the test during peak traffic hours vs quiet time. I typically do my code work in the quiet time so my test in the morning typically better than mid day or evening time when busy. Timing wise is tough to work with growing traffic and times get longer. As for the CLS you can fix that for the most part but adsense ads tend to create huge amount of CLS. 

Posted

Like currently to gain more speed and reduce timings here. I've now downloaded a copy of my entire website. Now taking my attachments which are mostly photos. Copied again on my local hard drive and going month by month to re-compress all the photos down. 

cd ..
cd monthly_2021_02
jpegoptim -f *.jpg
optipng *.png

In doing so I've seen several files with as high as 30% saving in space now. Just the month_2021_02 I made at least a 25 MB space savings. When you run through all the folders there might be a considerable saving of all the photos being re-compressed the marginal pages I've got will most likely pass when I get this project done. Now figure my website started way back at 2007 orignally so there is a close to 8.5 GB of files on the attachment folder and going to take me awhile to process down.

SNIP - jpegoptim -f *.jpg

0190327_134444.jpg.015a98b6afa7588b1fde06ae3c81cfe7.jpg 1600x1600 24bit N JFIF  [OK] 290807 --> 286312 bytes (1.55%), optimized.
20190327_135422.jpg.0eef6c35371e5728900cbd03ccf04d05.jpg 1600x1600 24bit N JFIF  [OK] 144554 --> 128616 bytes (11.03%), optimized.
20190327_135548.jpg.0537f4829d55748717ea4c663efdca4a.jpg 1600x1600 24bit N JFIF  [OK] 158866 --> 142351 bytes (10.40%), optimized.
20190327_135829.jpg.c58ea22e15835c0a5aa7856dbb2e603c.jpg 1600x1600 24bit N JFIF  [OK] 240411 --> 233764 bytes (2.76%), optimized.
20190327_180549.jpg.0721a2f8be87ab7b37784317744b5950.jpg 1200x1600 24bit N JFIF  [OK] 216775 --> 212396 bytes (2.02%), optimized.
20190327_190557.jpg.b005c88e1c935ad0f004ba0f7f6a1421.jpg 1600x1600 24bit N JFIF  [OK] 263128 --> 255889 bytes (2.75%), optimized.
20190327_190916.jpg.3f05b7cd72091ab793c11e147fe6a7e1.jpg 1600x1600 24bit N JFIF  [OK] 283643 --> 277559 bytes (2.14%), optimized.

SNIP - optipng *.png

1600x964 pixels, 4x8 bits/pixel, RGB+alpha
Reducing image to 3x8 bits/pixel, RGB
Input IDAT size = 565897 bytes
Input file size = 566803 bytes

Trying:
  zc = 9  zm = 8  zs = 0  f = 0		IDAT size = 394146
                               
Selecting parameters:
  zc = 9  zm = 8  zs = 0  f = 0		IDAT size = 394146

Output IDAT size = 394146 bytes (171751 bytes decrease)
Output file size = 394224 bytes (172579 bytes = 30.45% decrease)

 

Posted
2 hours ago, Jimi Wikman said:

Don't stare too much at the speed of the site or the rankings.

That's what I did for 5 years... Losing traffic and search rankings dropped. Now I'm focused on speed and meeting Google requirements and gain over 20,000 people per month and still growing. Traffic is improving and speed is getting slower because the server is getting busier. So what I did in the last 3 month netted me more than leaving it alone for 5 years and watched it fall apart. Better your speed ranking the better your ranking of SEO with Google.

Posted (edited)
5 hours ago, Mopar1973Man said:

 Now I'm focused on speed and meeting Google requirements and gain over 20,000 people per month and still growing. .

That’s great, but how do you know that your actions in regard to speed are what improves your ranking and therefore traffic?
Just because one happens after the other doesn’t mean much. That could just be the post hoc ergo propter hoc fallacy. 

Edited by opentype
Posted

Well they won't put you on news or discover without CWV scores being high. Discover and news drive insane traffic and are viewable from your google dashboard, also if you had google analytics or any other statistics it would be pretty easy to see the ingress from search and not from direct. 

Are you implying that all of the sudden his content got more useful? I'm pretty sure he's implying that a lot of content that got deranked is now top hit ranked. 

  • Recently Browsing   0 members

    • No registered users viewing this page.
×
×
  • Create New...