You can write algorithms in any language you want. Making them work is another matter.
It's pretty safe to say that National Instruments LabView, T/SQL, and GWBasic are radically different from each other but still follow fundamental rules. All have arrays, bit fields, string data types, floating point, etc. However, each is fit for a particular purpose, an in that respect language matters a lot.
It's possible 'language doesn't matter' if one is discussing the difference between Java, Perl, C#, and Ruby and most of your code is making round trips to the database for queries and updates. I would like to see you sample digital data at 80,000 samples per second in either T-SQL or GWBasic, and I would like to see how much code you've written in the language without invoking external packages or libraries. And I'm a little curious how you might do that in the GWBasic 64K memory constraint.
There are three (at least) axis of optimization for languages: programmer productivity, machine efficiency, and real world interface. Using T-SQL, I can create an expression in one line that can do a very complex transform on hundreds of millions of records out of a collection of billions. I don't spend any time thinking about how the machine does it, I write my expression and hit the green arrow. Done. One thing that tends to be true of databases is that the contents is usually 'clean', therefore I don't worry about whether a date is actually a date or a floating point is really a number.
In C, programmer productivity takes a back seat to machine constraints. If I have to 'make it fit' in 32K, and that 32K is on an embedded controller that's going to be spinning inside a tire on the freeway at 85 MPH at high noon on the 4th of July, I can't really tell the customer to add more memory if it runs out. If it takes me a year to write and test 2500 lines of code, so it goes. If that code goes into a billion tires the cost of that year of development is beneath notice.
Data types, data structures, and flow of control expressions are pretty much universal in languages. Gritty details about what happens in the real world are something else entirely.